JP7527758B2 - CD80 variant immunomodulatory proteins and uses thereof - Google Patents

Description

CROSS-REFERENCE TO RELATED APPLICATIONS This application is a continuation of U.S. Provisional Patent Application No. 62/353,595, filed April 15, 2016, entitled "CD80 Ligand Variant Immunomodulatory Proteins and Uses Thereof," U.S. Provisional Patent Application No. 62/394,743, filed September 14, 2016, entitled "CD80 Ligand Variant Immunomodulatory Proteins and Uses Thereof," U.S. Provisional Patent Application No. 62/410,844, filed October 20, 2016, entitled "CD80 Ligand Variant Immunomodulatory Proteins and Uses Thereof," U.S. Provisional Patent Application No. 62/410,844, filed March 16, 2017, entitled "CD80 Ligand Variant Immunomodulatory Proteins and Uses Thereof," No. 62/472,570, entitled "CD80 Ligand Variant Immunomodulatory Proteins and Uses Thereof," filed March 22, 2017, and U.S. Provisional Patent Application No. 62/475,201, entitled "CD80 Ligand Variant Immunomodulatory Proteins and Uses Thereof," filed March 22, 2017, the contents of each of which are incorporated by reference in their entirety.

INCORPORATION BY REFERENCE TO SEQUENCE LISTING This application is filed together with a Sequence Listing in electronic format. The Sequence Listing is provided as a file entitled 761612000440SeqList.txt, created on April 13, 2017, and having a size of 815,991 bytes. The information in the electronic format of the Sequence Listing is incorporated by reference in its entirety.

FIELD The present disclosure relates to therapeutic compositions for modulating immune responses in the treatment of cancer and immunological diseases. In some aspects, the present disclosure relates to specific variants of CD80 that exhibit altered binding affinity to their cognate binding partners.

Background There is growing medical interest in modulating immune responses by intervening in processes occurring at the immune synapse (IS) formed between antigen-presenting cells (APCs) or target cells and lymphocytes. Mechanistically, cell surface proteins within the IS may be involved in coordinated and often simultaneous interactions between multiple protein targets and a single protein to which they bind. Interactions at the IS occur in close association with the junction of two cells, and a single protein within this structure can interact (possibly simultaneously) with proteins both on the same cell (cis) and on interacting cells (trans). Although therapeutic agents capable of modulating the IS are known, improved therapeutic agents are needed. Provided are immunomodulatory proteins, including soluble or transmembrane immunomodulatory proteins that can be expressed on cells that meet this need.

SUMMARY In some embodiments, a variant CD80 polypeptide is provided that contains an IgV domain or a specific binding fragment thereof, an IgC domain or a specific binding fragment thereof, or both, and is represented by SEQ ID NO: NO: Position based on 28 numbering 4, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 20, 21, 22, 24, 25, 27, 28, 29, 30, 31, 33, 36, 37, 38, 40, 41, 42, 43, 44, 47, 48, 50, 52, 53, 54, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 72, 74, 76, 77, 80, 81, 83, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 99, 102, 103, 104, 108, 107, 1 Provided herein are variant CD80 polypeptides containing one or more amino acid modifications in unmodified CD80 or a specific binding fragment thereof corresponding to: 09, 110, 114, 115, 116, 117, 118, 120, 121, 122, 126, 127, 128, 129, 130, 133, 137, 140, 142, 143, 144, 148, 149, 152, 154, 160, 162, 164, 168, 169, 174, 175, 177, 178, 183, 178, 185, 188, 190, 192, 193, or 199. In some embodiments, the one or more amino acid modifications comprise one or more amino acid substitutions, insertions, or deletions. In some embodiments, the unmodified CD80 is a mammalian CD80 or a specific binding fragment thereof. In some embodiments, the unmodified CD80 is human CD80 or a specific binding fragment thereof. In some embodiments, the variant CD80 polypeptide comprises an IgV domain, or a specific binding fragment thereof; and an IgC domain, or a specific binding fragment thereof. In some embodiments, the unmodified CD80 comprises (i) an amino acid sequence as set forth in SEQ ID NO:28, (ii) an amino acid sequence having at least 95% sequence identity to SEQ ID NO:28; or (iii) a portion thereof, which comprises an IgV domain or an IgC domain, or a specific binding fragment thereof.

In some embodiments of any one of the variant CD80 polypeptides, the specific binding fragment of the IgV domain or the IgC domain has a length of at least 50, 60, 70, 80, 90, 100, 110 or more amino acids; the specific binding fragment of the IgV domain comprises at least 80% of the length of the IgV domain set forth as amino acids 35-135 of SEQ ID NO:1; or the specific binding fragment of the IgC domain comprises at least 80% of the length of the IgC domain set forth as amino acids 145-230 of SEQ ID NO:1. In some embodiments, the variant CD80 polypeptide contains up to 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20 amino acid modifications (optionally amino acid substitutions). In some embodiments, the variant CD80 polypeptide comprises an amino acid sequence that exhibits at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity to SEQ ID NO:28 or a specific binding fragment thereof. In some embodiments, the variant CD80 has altered binding specificity for CD28, PD-L1, or CTLA-4 ectodomain compared to unmodified CD80. In some cases, the altered binding is altered binding affinity and/or altered binding selectivity.

またはその保存的アミノ酸置換である。 In some embodiments of any one of the variant CD80 polypeptides, the one or more amino acid substitutions are

or a conservative amino acid substitution thereof.

In some embodiments, the variant CD80 polypeptide contains an IgV or IgC domain or a specific binding fragment thereof that contains one or more amino acid deletions in unmodified CD80 or a specific binding fragment thereof. In some embodiments, the deletion corresponds to position 43 of SEQ ID NO:28.

である。 In some embodiments, the one or more amino acid substitutions are:

It is.

In some embodiments of any one of the variant CD80 polypeptides, the variant CD80 polypeptide contains an IgV domain or a specific fragment thereof and/or an IgC domain or a specific fragment thereof. In some embodiments of any one of the variant CD80 polypeptides, the variant CD80 polypeptide contains an IgV domain or a specific binding fragment thereof. In some embodiments, the IgV domain or a specific binding fragment thereof is the only CD80 portion of the variant CD80 polypeptide. In some embodiments, the IgC domain or a specific binding fragment thereof is the only CD80 portion of the variant CD80 polypeptide.

In some embodiments of any one of the variant CD80 polypeptides, the variant CD80 polypeptide comprises an amino acid sequence set forth in any of SEQ ID NOs:55-108, 280-346, 414-475, or a specific binding fragment thereof, or an amino acid sequence that exhibits at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% sequence identity to any of SEQ ID NOs:55-108, 280-346, 414-475, or a specific binding fragment thereof and contains one or more amino acid substitutions. In some embodiments, the variant CD80 polypeptide comprises an IgV domain or a specific binding fragment thereof. In some embodiments, the IgV domain or a specific binding fragment thereof is the only CD80 portion of the variant CD80 polypeptide. In some embodiments, the IgC domain or specific fragment thereof is the only CD80 portion of the variant CD80 polypeptide. In some embodiments of any one of the variant CD80 polypeptides, the variant CD80 polypeptide comprises an amino acid sequence set forth in any one of SEQ ID NOs: 153-195, 347, 373-386, 476-477, or a specific binding fragment thereof, an amino acid sequence that exhibits at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% sequence identity to any one of SEQ ID NOs: 153-195, 347, 373-386, 476-477, or a specific binding fragment thereof, and contains one or more amino acid substitutions.

In some embodiments of any one of the variant CD80 polypeptides, the variant CD80 polypeptide specifically binds to the ectodomain of CD28, PD-L1, or CTLA-4 with improved affinity compared to the unmodified CD80 polypeptide. In some of any such embodiments, the variant polypeptide specifically binds to the ectodomain of CD28, PD-L1, or CTLA-4 with improved selectivity compared to the binding of unmodified CD80 to the ectodomain. In some aspects, the improved selectivity includes a greater ratio of binding to one cognate binding partner selected from among CD28, PD-L1, and CTLA-4 and binding to another of the cognate binding partners compared to the ratio of binding of the unmodified CD80 polypeptide to the one cognate binding partner and binding of the unmodified CD80 polypeptide to the other of the cognate binding partners.

In some embodiments, the variant polypeptides specifically bind to the ectodomain of CD28 with improved selectivity as compared to the binding of unmodified CD80 to the ectodomain of CD28. In some aspects, the improved selectivity includes a greater ratio of binding to CD28 to PD-L1 or CTLA-4 as compared to the ratio of binding of the unmodified CD80 polypeptide to CD28 to PD-L1 or CTLA-4. In some cases, the ratio is at least or at least about 1.5-fold, 2.0-fold, 3.0-fold, 4.0-fold, 5-fold, 10-fold, 15-fold, 20-fold, 30-fold, 40-fold, 50-fold, or more.

In some embodiments, the variant CD80 polypeptide specifically binds to the ectodomain of CD28 with improved affinity compared to an unmodified CD80 polypeptide. In some embodiments, the improved affinity for the ectodomain is greater than 1.2-fold, 1.5-fold, 2-fold, 3-fold, 4-fold, 5-fold, 6-fold, 7-fold, 8-fold, 9-fold, 10-fold, 20-fold, 30-fold, 40-fold, 50-fold, or 60-fold improved compared to the binding affinity of unmodified CD80 for the ectodomain.

ならびにその保存的アミノ酸置換からなる群より選択される。いくつかの態様において、1つまたは複数のアミノ酸置換は、

である。 In some embodiments of any one of the variant CD80 polypeptides, the one or more amino acid substitutions correspond to positions 12, 18, 20, 29, 31, 36, 40, 41, 43, 52, 59, 60, 63, 67, 70, 77, 81, 87, 88, 89, 90, 91, 92, 93, 107, 109, 114, 117, 118, 120, 122, 127, 130, 144, 169, 177, or 199 based on the numbering of SEQ ID NO:28.

and conservative amino acid substitutions thereof. In some embodiments, the one or more amino acid substitutions are selected from the group consisting of:

It is.

In some embodiments of any one of the variant CD80 polypeptides, the variant CD80 polypeptide specifically binds to the ectodomain of PD-L1 with improved affinity compared to an unmodified CD80 polypeptide. In some cases, the improved affinity for the ectodomain is greater than 1.2-fold, 1.5-fold, 2-fold, 3-fold, 4-fold, 5-fold, 6-fold, 7-fold, 8-fold, 9-fold, 10-fold, 20-fold, 30-fold, 40-fold, 50-fold, or 60-fold improved compared to the binding affinity of unmodified CD80 for the ectodomain. In some of any such embodiments, the variant polypeptide specifically binds to the ectodomain of PD-L1 with improved selectivity compared to the binding affinity of unmodified CD80 for the ectodomain. In some cases, the improved selectivity includes a ratio of binding to PD-L1 to binding to CD28 or CTLA-4 that is greater than the ratio of binding of the unmodified CD80 polypeptide to PD-L1 to binding of the unmodified CD80 polypeptide to CD28 or CTLA-4. In some cases, the ratio is at least or at least about 1.5-fold, 2.0-fold, 3.0-fold, 4.0-fold, 5-fold, 10-fold, 15-fold, 20-fold, 30-fold, 40-fold, 50-fold, or more greater.

ならびにその保存的アミノ酸置換からなる群より選択される。いくつかの態様において、1つまたは複数のアミノ酸置換は、

である。 In some embodiments, the one or more amino acid substitutions correspond to positions 12, 18, 29, 31, 33, 36, 38, 40, 41, 42, 43, 47, 59, 67, 70, 77, 81, 85, 87, 88, 89, 90, 91, 92, 93, 94, 109, 118, 120, 122, 144, 148, 149, or 177 based on the numbering of SEQ ID NO:28.

and conservative amino acid substitutions thereof. In some embodiments, the one or more amino acid substitutions are selected from the group consisting of:

It is.

およびその保存的アミノ酸置換からなる群より選択される。いくつかの態様において、1つまたは複数のアミノ酸置換は、

である。 In some embodiments of any one of the variant CD80 polypeptides, the variant CD80 polypeptide specifically binds to the ectodomain of CD28 and the ectodomain of PD-L1 with improved affinity compared to an unmodified CD80 polypeptide. In some embodiments, the one or more amino acid substitutions correspond to positions 12, 18, 36, 40, 43, 59, 77, 88, 89, 90, 91, 92, 93, 109, 118, 122, 177 based on the numbering of SEQ ID NO:28 ...

and conservative amino acid substitutions thereof. In some embodiments, the one or more amino acid substitutions are selected from the group consisting of:

It is.

In some embodiments of any one of the variant CD80 polypeptides, the variant CD80 polypeptide specifically binds to the ectodomain of CTLA-4 with improved affinity compared to the unmodified CD80 polypeptide. In some embodiments, the improved affinity for the ectodomain is 1.2-fold, 1.5-fold, 2-fold, 3-fold, 4-fold, 5-fold, 6-fold, 7-fold, 8-fold, 9-fold, 10-fold, 20-fold, 30-fold, 40-fold, 50-fold, or 60-fold or more improved compared to the binding affinity of unmodified CD80 for the ectodomain. In some of any such embodiments, the variant polypeptide specifically binds to the ectodomain of CTLA-4 with improved selectivity compared to the binding affinity of unmodified CD80 for the ectodomain. In some cases, the improved selectivity includes a greater ratio of binding to CTLA-4 to binding to CD28 or PD-L1 compared to the ratio of binding of the unmodified CD80 polypeptide to CTLA-4 to binding to CD28 or PD-L1. In some embodiments, the ratio is at least or at least about 1.5-fold, 2.0-fold, 3.0-fold, 4.0-fold, 5-fold, 10-fold, 15-fold, 20-fold, 30-fold, 40-fold, 50-fold, or more greater.

ならびにその保存的アミノ酸置換からなる群より選択される。いくつかの態様において、1つまたは複数のアミノ酸置換は、

である。 In some embodiments, the one or more amino acid substitutions correspond to positions 4, 29, 31, 36, 40, 41, 52, 67, 68, 70, 87, 88, 89, 90, 91, 92, 93, 107, 109, 110, 118, 120, 130, 144, or 169 based on the numbering of SEQ ID NO:28.

and conservative amino acid substitutions thereof. In some embodiments, the one or more amino acid substitutions are selected from the group consisting of:

It is.

ならびにその保存的アミノ酸置換からなる群より選択される。いくつかの態様において、1つまたは複数のアミノ酸置換は、

である。 In some embodiments of any one of the variant CD80 polypeptides, the variant CD80 polypeptide specifically binds to the ectodomain of CD28 and the ectodomain of CTLA-4 with improved affinity as compared to an unmodified CD80 polypeptide. In some embodiments, the one or more amino acid substitutions correspond to positions 36, 40, 52, 70, 88, 89, 90, 91, 92, 93, 107, 118, 120, 130, 144, or 169 of SEQ ID NO:28 ...

and conservative amino acid substitutions thereof. In some embodiments, the one or more amino acid substitutions are selected from the group consisting of:

It is.

およびその保存的アミノ酸置換からなる群より選択される。いくつかの態様において、1つまたは複数のアミノ酸置換は、

である。 In some embodiments of any one of the variant CD80 polypeptides, the variant CD80 polypeptide specifically binds to the ectodomain of PD-L1 and the ectodomain of CTLA-4 with improved affinity compared to an unmodified CD80 polypeptide. In some embodiments, the one or more amino acid substitutions correspond to positions 29, 31, 36, 40, 41, 67, 70, 87, 88, 89, 90, 91, 92, 93, 109, 118, 120, 122, or 178 of SEQ ID NO:28 ...

and conservative amino acid substitutions thereof. In some embodiments, the one or more amino acid substitutions are selected from the group consisting of:

It is.

である。 In some embodiments of any one of the variant CD80 polypeptides, the variant CD80 polypeptide specifically binds to the ectodomain of CD28, the ectodomain of PD-L1, and the ectodomain of CTLA-4 with improved affinity as compared to an unmodified CD80 polypeptide. In some embodiments, the one or more amino acid substitutions correspond to positions 36, 40, 88, 89, 90, 91, 92, or 93 of SEQ ID NO:28. In some embodiments, the one or more amino acid substitutions are selected from the group consisting of K36G, L40M, E88D, K89R, D90K, A91G, F92Y, K93R, and conservative amino acid substitutions thereof. In some embodiments, the one or more amino acid substitutions are

It is.

ならびにその保存的アミノ酸置換からなる群より選択される。いくつかの態様において、1つまたは複数のアミノ酸置換は、

である。 In some embodiments of any one of the variant CD80 polypeptides, the variant CD80 polypeptide specifically binds to the ectodomain of CD28 or the ectodomain of PD-L1 with improved affinity compared to an unmodified CD80 polypeptide, and the variant CD80 polypeptide specifically binds to the ectodomain of CTLA-4 with reduced affinity compared to an unmodified CD80 polypeptide. In some embodiments, the one or more amino acid substitutions correspond to positions 29, 31, 33, 36, 38, 41, 42, 43, 47, 63, 67, 70, 81, 85, 87, 88, 89, 90, 91, 92, 93, 94, 109, 114, 118, 120, 127, 130, 144, 148, or 149 of SEQ ID NO:28 ...

and conservative amino acid substitutions thereof. In some embodiments, the one or more amino acid substitutions are selected from the group consisting of:

It is.

である。 In some embodiments of any one of the variant CD80 polypeptides, the variant CD80 polypeptide specifically binds to the ectodomain of CD28 with improved affinity as compared to an unmodified CD80 polypeptide, and the variant CD80 polypeptide specifically binds to the ectodomain of CTLA-4 with reduced affinity as compared to an unmodified CD80 polypeptide. In some embodiments, the one or more amino acid substitutions correspond to positions 63, 70, 81, 87, 91, 114, 118, 120, 127, or 130 of SEQ ID NO:28. In some embodiments, the one or more amino acid substitutions are selected from the group consisting of N63S, L70Q, E81A, Y87N, A91G, S114T, I118V, T120S, I127T, and T130A, and conservative amino acid substitutions thereof. In some embodiments, the one or more amino acid substitutions are

It is.

ならびにその保存的アミノ酸置換からなる群より選択される。いくつかの態様において、1つまたは複数のアミノ酸置換は、

である。 In some embodiments of any one of the variant CD80 polypeptides, the variant CD80 polypeptide specifically binds to the ectodomain of PD-L1 with improved affinity compared to an unmodified CD80 polypeptide, and the variant CD80 polypeptide specifically binds to the ectodomain of CTLA-4 with reduced affinity compared to an unmodified CD80 polypeptide. In some embodiments, the one or more amino acid substitutions correspond to positions 29, 31, 33, 36, 38, 41, 42, 43, 47, 67, 70, 81, 85, 87, 88, 89, 90, 91, 92, 93, 94, 109, 118, 120, 144, 148, 149 of SEQ ID NO:28 ...

and conservative amino acid substitutions thereof. In some embodiments, the one or more amino acid substitutions are selected from the group consisting of:

It is.

In some embodiments of any one of the variant CD80 polypeptides, the variant CD80 polypeptide specifically binds to the ectodomain of CD28 and the ectodomain of PD-L1 with improved affinity compared to the unmodified CD80 polypeptide, and the variant CD80 polypeptide specifically binds to the ectodomain of CTLA-4 with reduced affinity compared to the unmodified CD80 polypeptide. In some embodiments, the one or more amino acid substitutions correspond to positions 70, 81, 87, 91, or 120 of SEQ ID NO:28. In some embodiments, the one or more amino acid substitutions are selected from the group consisting of L70Q, Y87N, A91G, and T120S, and conservative amino acid substitutions thereof.

In some embodiments of any one of the variant CD80 polypeptides, CD28 is human CD28. In some embodiments, PD-L1 is human PD-L1. In some embodiments, CTLA-4 is human CTLA-4.

In some embodiments of any one of the variant CD80 polypeptides, the binding activity is altered (improved or decreased) by more than 1.2-fold, 1.5-fold, 2-fold, 3-fold, 4-fold, 5-fold, 6-fold, 7-fold, 8-fold, 9-fold, 10-fold, 20-fold, 30-fold, 40-fold, or 50-fold compared to the unmodified CD80 polypeptide.

In some embodiments of any one of the variant CD80 polypeptides, the variant CD80 polypeptide is a soluble protein.

In some embodiments of any one of the variant CD80 polypeptides, the variant CD80 polypeptide is linked to a multimerization domain. In some embodiments, the variant CD80 polypeptide is a multimeric polypeptide, optionally a dimeric polypeptide, comprising a first variant CD80 polypeptide linked to a multimerization domain and a second variant CD80 polypeptide linked to a multimerization domain. In some embodiments, the first variant CD80 polypeptide and the second variant CD80 polypeptide are the same or different. In some embodiments, the multimerization domain is an Fc domain or a variant thereof having reduced effector function.

In some embodiments of any one of the variant CD80 polypeptides, the variant CD80 polypeptide is linked to a moiety that increases the biological half-life of the polypeptide. In some embodiments, the variant CD80 polypeptide is linked to an Fc domain or variant thereof that has reduced effector function. In some embodiments, the Fc domain is mammalian (optionally human); or the variant Fc domain contains one or more amino acid modifications compared to an unmodified Fc domain that is mammalian (optionally human). In some embodiments, the Fc domain or variant thereof contains an amino acid sequence set forth in SEQ ID NO:226 or SEQ ID NO:227, or contains an amino acid sequence that exhibits at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% sequence identity to SEQ ID NO:226 or SEQ ID NO:227. In some embodiments, the variant CD80 polypeptide is indirectly linked via a linker.

In some embodiments of any one of the variant CD80 polypeptides that are transmembrane immunomodulatory proteins, the variant CD80 polypeptide further contains a transmembrane domain linked to the extracellular domain (ECD) of the variant CD80 polypeptide or a specific binding fragment thereof. In some embodiments, the transmembrane domain contains the amino acid sequence set forth as residues 243-263 of SEQ ID NO:1, or a functional variant thereof that exhibits at least 85% sequence identity to residues 243-263 of SEQ ID NO:1. In some embodiments, the variant CD80 polypeptide further contains a cytoplasmic signaling domain linked to the transmembrane domain. In some embodiments, the cytoplasmic signaling domain contains an amino acid sequence set forth as residues 264-288 of SEQ ID NO:1, or a functional variant thereof that exhibits at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% sequence identity to residues 264-288 of SEQ ID NO:1.

In some embodiments of any one of the variant CD80 polypeptides, the variant CD80 increases IFN-γ (interferon-gamma) expression in an in vitro primary T cell assay relative to unmodified CD80. In some embodiments of any one of the variant CD80 polypeptides, the variant CD80 decreases IFN-γ (interferon-gamma) expression in an in vitro primary T cell assay relative to unmodified CD80. In some embodiments of any one of the variant CD80 polypeptides, the variant CD80 polypeptide is deglycosylated.

In some embodiments, provided herein is an immune modulating polypeptide comprising a variant CD80 according to any one of the embodiments described herein linked to a second polypeptide comprising an immunoglobulin superfamily (IgSF) domain. In some embodiments, the IgSF domain has modified affinity and altered binding to one or more of its cognate binding partners compared to an unmodified or wild-type IgSF domain. In some embodiments, the IgSF domain has improved binding to one or more of its cognate binding partners compared to an unmodified or wild-type IgSF domain. In some embodiments, the variant CD80 is a first variant CD80 and the IgSF domain of the second polypeptide is an IgSF domain from a second variant CD80 according to any one of the embodiments described herein, wherein the first and second variant CD80 are the same or different. In some embodiments, the variant CD80 polypeptide is capable of specifically binding to CD28, PD-L1, or CTLA-4, and the IgSF domain of the second polypeptide is capable of binding to a cognate binding partner other than that to which the variant CD80 polypeptide specifically binds. In some embodiments, the variant CD80 polypeptide is capable of specifically binding to CD28 or PD-L1, and the IgSF domain is capable of binding to a cognate binding partner other than that to which the variant CD80 polypeptide specifically binds. In some embodiments, the variant CD80 polypeptide is capable of specifically binding to CD28 or CTLA-4, and the IgSF domain is capable of binding to a cognate binding partner other than that to which the variant CD80 polypeptide specifically binds. In some embodiments, the variant CD80 polypeptide is capable of specifically binding to CTLA-4 or PD-L1, and the IgSF domain is capable of binding to a cognate binding partner other than that to which the variant CD80 polypeptide specifically binds. In some embodiments, the IgSF domain is from a member of the B7 family. In some embodiments, the IgSF domain is a tumor-localizing moiety that binds to a ligand expressed on tumors. In some embodiments, the ligand is B7H6. In some embodiments, the IgSF domain is from NKp30. In some embodiments, the IgSF domain has modified affinity and improved binding to one or more of its cognate binding partners compared to an unmodified or wild-type IgSF domain. In some embodiments, the IgSF domain is or contains an IgV domain. In some embodiments, the variant CD80 polypeptide is or contains an IgV domain.

In some embodiments of any one of the immunomodulatory proteins, the immunomodulatory protein contains a multimerization domain linked to one or both of a variant CD80 polypeptide and a second polypeptide (including an IgSF domain). In some embodiments, the multimerization domain is an Fc domain or a variant thereof having reduced effector function. In some embodiments, the immunomodulatory protein is a dimer. In some embodiments, the immunomodulatory protein is a homodimer. In some cases, the immunomodulatory protein is a heterodimer.

In some embodiments, provided herein is a conjugate comprising a variant CD80 according to any one of the embodiments described herein or an immune modulating polypeptide according to any one of the embodiments described herein linked to a moiety. In some embodiments, the moiety is a targeting moiety that specifically binds to a molecule on the surface of a cell. In some embodiments, the targeting moiety specifically binds to a molecule on the surface of an immune cell. In some embodiments, the immune cell is an antigen presenting cell or a lymphocyte. In some embodiments, the targeting moiety is a tumor-localizing moiety that binds to a molecule on the surface of a tumor. In some embodiments, the moiety is a protein, peptide, nucleic acid, small molecule, or nanoparticle. In some embodiments, the moiety is an antibody or antigen-binding fragment. In some of any such embodiments, the conjugate is bivalent, tetravalent, hexavalent, or octavalent.

In some embodiments, provided herein is a nucleic acid molecule encoding a variant CD80 according to any one of the embodiments described herein or an immune modulating polypeptide according to any one of the embodiments described herein. In some embodiments, the nucleic acid molecule is a synthetic nucleic acid. In some embodiments, the nucleic acid is a cDNA.

In some embodiments, provided herein is a vector comprising a nucleic acid of any one of the embodiments described herein. In some embodiments, the vector is an expression vector.

In some embodiments, the vector is a mammalian vector or a viral vector.

In some embodiments, provided herein is a cell comprising a vector according to any one of the embodiments described herein. In some embodiments, the cell is a mammalian cell. In some embodiments, the cell is a human cell.

In some embodiments, provided herein is a method of producing a variant CD80 polypeptide or immunomodulatory protein, the method comprising introducing a nucleic acid molecule according to any one of the embodiments described herein or a vector according to any one of the embodiments described herein into a host cell under conditions that allow expression of the protein in the cell. In some embodiments, the method further comprises isolating or purifying the variant CD80 polypeptide or immunomodulatory protein from the cell.

In some embodiments, provided herein is a method of modifying a cell to express a variant CD80 polypeptide, the method comprising introducing into a host cell a nucleic acid molecule encoding a variant CD80 polypeptide according to any one of the embodiments described herein under conditions in which the polypeptide is expressed in the cell.

In some embodiments, provided herein is a modified cell expressing a variant CD80 polypeptide according to any one of the embodiments described herein, an immunomodulatory protein according to any one of the embodiments described herein, a nucleic acid molecule according to any one of the embodiments described herein, or a vector according to any one of the embodiments described herein. In some embodiments, the variant CD80 polypeptide or immunomodulatory polypeptide contains a signal peptide. In some aspects, the variant CD80 polypeptide or immunomodulatory polypeptide does not contain a transmembrane domain and/or is not expressed on the surface of the cell. In some embodiments, the variant CD80 polypeptide or immunomodulatory polypeptide is secreted from the modified cell.

In some embodiments, the modified cell contains a variant CD80 polypeptide that contains a transmembrane domain and/or is a transmembrane immunomodulatory protein according to any one of the embodiments described herein. In some embodiments, the variant CD80 polypeptide is expressed on the surface of the cell.

In some embodiments, the modified cell is an immune cell. In some embodiments, the immune cell is an antigen presenting cell (APC) or a lymphocyte. In some embodiments, the modified cell is a primary cell. In some embodiments, the modified cell is a mammalian cell. In some embodiments, the modified cell is a human cell. In some embodiments, the lymphocyte is a T cell. In some embodiments, the APC is an artificial APC.

In some of any such embodiments, the modified cells further contain a chimeric antigen receptor (CAR) or a modified T cell receptor.

Also provided is an infectious agent comprising a nucleic acid molecule encoding a variant CD80 polypeptide according to any one of the embodiments described herein or an immunomodulatory polypeptide according to any one of the embodiments described herein. In some cases, the encoded variant CD80 polypeptide or immunomodulatory polypeptide does not contain a transmembrane domain and/or is not expressed on the surface of the cell in which it is expressed. In some embodiments, the encoded variant CD80 polypeptide or immunomodulatory polypeptide is secreted from the cell in which it is expressed. In some aspects, the encoded variant CD80 polypeptide contains a transmembrane domain. In some aspects, the encoded variant CD80 polypeptide is expressed on the surface of the cell in which it is expressed.

In some of any such embodiments, the infectious agent is a bacterium or a virus. In some cases, the virus is an oncolytic virus. In some embodiments, the oncolytic virus is an adenovirus, an adeno-associated virus, a herpes virus, a herpes simplex virus, a vesicular stomatic virus, a reovirus, a Newcastle disease virus, a parvovirus, a measles virus, a vesicular stomatitis virus (VSV), a coxsackievirus, or a vaccinia virus. In some embodiments, the virus specifically targets and/or is dendritic cell tropic to dendritic cells (DCs). In some cases, the virus is a lentiviral vector pseudotyped with a modified Sindbis virus envelope product.

In some of any such embodiments, the infectious agent further contains a nucleic acid molecule encoding an additional gene product that can cause death of a target cell or enhance or potentiate an immune response. In some embodiments, the additional gene product is selected from an anti-cancer agent, an anti-metastatic agent, an anti-angiogenic agent, an immunomodulatory molecule, an immune checkpoint inhibitor, an antibody, a cytokine, a growth factor, an antigen, a cytotoxic gene product, a pro-apoptotic gene product, an anti-apoptotic gene product, a cell matrix degrading gene, a gene for tissue regeneration, or a gene for reprogramming a human somatic cell to pluripotency.

In some embodiments, provided herein is a pharmaceutical composition comprising a variant CD80 polypeptide according to any one of the embodiments described herein, an immunomodulatory protein according to any one of the embodiments described herein, a conjugate according to any one of the embodiments described herein, or a modified cell according to any one of the embodiments described herein. In some embodiments, the pharmaceutical composition further comprises a pharma- ceutically acceptable excipient. In some embodiments, the pharmaceutical composition is sterile.

In some embodiments, provided herein is an article of manufacture comprising a pharmaceutical composition according to any one of the embodiments described herein in a vial. In some embodiments, the vial is sealed.

In some embodiments, provided herein is a kit comprising a pharmaceutical composition according to any one of the embodiments described herein and instructions for use. In some embodiments, provided herein is a kit comprising an article of manufacture according to any one of the embodiments described herein and instructions for use.

In some embodiments, provided herein is a method of modulating an immune response in a subject, the method comprising administering to the subject a pharmaceutical composition according to any one of the embodiments described herein. In some embodiments, the method comprises administering to the subject a modified cell according to any one of the embodiments described herein. In some embodiments, the modified cell is autologous to the subject. In some embodiments, the modified cell is allogeneic to the subject.

In some embodiments, modulating the immune response treats a disease or condition in a subject. In some embodiments, the immune response is enhanced. In some embodiments, an immunomodulating protein or conjugate comprising a variant CD80 polypeptide linked to a tumor-localizing moiety is administered to a subject. In some cases, the tumor-localizing moiety is or contains a binding molecule that recognizes a tumor antigen. In some cases, the binding molecule contains an antibody or antigen-binding fragment thereof, or contains a wild-type IgSF domain or a variant thereof.

In some embodiments, a pharmaceutical composition comprising an immunomodulatory protein according to any one of the embodiments described herein or a conjugate according to any one of the embodiments described herein is administered to the subject. In some embodiments, a modified cell comprising a variant CD80 polypeptide that is a transmembrane immunomodulatory protein is administered to the subject and/or a modified cell according to any one of the embodiments described herein is administered to the subject.

In some embodiments, an infectious agent encoding a variant CD80 polypeptide that is a transmembrane immunomodulatory protein is administered to a subject, optionally under conditions in which the infectious agent infects tumor cells or immune cells and the transmembrane immunomodulatory protein is expressed on the surface of the infected cells. In some aspects, the transmembrane immunomodulatory protein is a transmembrane immunomodulatory protein according to any one of the embodiments described herein.

In some embodiments, the disease or condition is a tumor or cancer. In some embodiments, the disease or condition is selected from melanoma, lung cancer, bladder cancer, hematological malignancies, liver cancer, brain cancer, kidney cancer, breast cancer, pancreatic cancer, colon cancer, spleen cancer, prostate cancer, testicular cancer, ovarian cancer, uterine cancer, gastric cancer, musculoskeletal cancer, head and neck cancer, gastrointestinal cancer, germ cell cancer, or endocrine and neuroendocrine cancer.

In some embodiments, the immune response is reduced by the provided methods of modulating the immune response.

In some embodiments, a variant CD80 polypeptide or immunomodulatory protein is administered to the subject that is soluble. In some cases, the soluble polypeptide or immunomodulatory protein is an Fc fusion protein. In some of any such embodiments, a pharmaceutical composition is administered to the subject that includes a variant CD80 polypeptide according to any one of the embodiments described herein, or an immunomodulatory protein according to any one of the embodiments described herein. In some embodiments, a modified cell is administered to the subject that includes a secreted variant CD80 polypeptide. In some embodiments, a modified cell is administered to the subject that includes any one of the embodiments described herein.

In some embodiments, an infectious agent encoding a variant CD80 polypeptide that is a secreted immunomodulatory protein is administered to a subject, optionally under conditions where the infectious agent infects tumor cells or immune cells and the secreted immunomodulatory protein is secreted from the infected cells.

In some of any such embodiments, the disease or condition is an inflammatory or autoimmune disease or condition. In some embodiments, the disease or condition is antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis, vasculitis, an autoimmune skin disease, transplantation, a rheumatic disease, an inflammatory gastrointestinal disease, an inflammatory eye disease, an inflammatory nervous system disease, an inflammatory lung disease, an inflammatory endocrine disease, or an autoimmune hematological disease. In some embodiments, the disease or condition is selected from inflammatory bowel disease, transplant, Crohn's disease, ulcerative colitis, multiple sclerosis, asthma, rheumatoid arthritis, or psoriasis.

Figures 1A-1C show various formats of variant IgSF domain molecules provided. Figure 1A shows a soluble molecule including: (1) a variant IgSF domain (vIgD), e.g., variant CD80, fused to an Fc chain; (2) a stacked molecule containing a first variant IgSF domain (first vIgD), e.g., variant CD80, and a second IgSF domain, e.g., second variant IgSF domain (second vIgD); (3) a tumor-targeted IgSF molecule containing a first variant IgSF domain (vIgD), e.g., variant CD80, and an IgSF domain that targets a tumor antigen, e.g., NkP30 IgSF domain; and (4) a variant IgSF domain (vIgD), e.g., variant CD80, linked to an antibody (V-Mab). 1B shows a transmembrane immunomodulatory protein (TIP) containing a variant IgSF domain (vIgD) expressed on the surface of a cell. In an exemplary embodiment, the cognate binding partner of the transmembrane-bound vIgD is a costimulatory receptor, and the TIP containing vIgD (e.g., CD80 vIgD) activates the costimulatory receptor such that the TIP induces a positive signal in cells expressing the costimulatory receptor. FIG. 1C shows a secreted immunomodulatory protein (SIP) in which a variant IgSF domain (vIgD) is secreted from a cell, for example, a first T cell (e.g., a CAR T cell). In an exemplary embodiment, the cognate binding partner of the secreted vIgD is an activating receptor, for example, CD28, which can be expressed on a first cell (e.g., a T cell) and/or a second cell (e.g., a T cell; either endogenous or modified, for example, a CAR T cell). The binding of the SIP to its cognate binding partner blocks signaling through the activating receptor. In all cases, the vIgD can be simply a V-domain (IgV), or a combination of a V-domain (IgV) and a C-domain (IgC) including the entire extracellular domain (ECD), or any combination of Ig domains of IgSF superfamily members. 1 shows an exemplary schematic diagram of the activity of a variant IgSF domain (vIgD) fused to Fc (vIgD-Fc), where vIgD is a variant of the IgSF domain of CD80. As shown, soluble vIgD of CD80 interacts with its cognate binding partner and blocks the interaction of CD28, thereby blocking costimulation by the CD28 costimulatory receptor. 1 shows an exemplary schematic diagram of a stack molecule for localizing variant IgSF domain (vIgD) to tumor cells. In this format, the stack molecule contains a first variant IgSF domain (first vIgD) and a second IgSF domain (e.g., second vIgD), where the second IgSF domain (e.g., second vIgD) is a tumor-targeting IgSF domain that binds to tumor antigen. An exemplary tumor-targeting IgSF domain is the IgSF domain of NkP30, which binds to tumor antigen B7-H6. In this depiction, vIgD is a variant of the IgSF domain of CD80. As shown, binding of the tumor-targeting IgSF domain to the surface of tumor cells localizes the first vIgD on the tumor cell surface, where it can interact with one or more of its cognate binding partners (e.g., CD28) expressed on the surface of adjacent immune cells (e.g., T cells) to stimulate co-stimulatory receptors. Various exemplary configurations of stack molecules containing a first variant IgSF domain (first vIgD) and a second IgSF domain, such as a second variant IgSF domain (second vIgD), are shown. As shown, the first vIgD and the second IgSF domain are independently linked, directly or indirectly, to the N-terminus or C-terminus of the Fc subunit. To generate homodimeric Fc molecules, the Fc subunits are matched and can form homodimers by co-expression of individual Fc subunits in a cell. To generate heterodimeric Fc molecules, the individual Fc subunits contain mutations (e.g., "knob-into-hole" mutations in the CH3 domain) that favor the formation of heterodimers compared to homodimers when the individual Fc subunits are co-expressed in a cell. 1 shows an exemplary schematic diagram of the activity of variant IgSF domain (vIgD) conjugated to antibody (V-Mab), in which the antibody (e.g., anti-HER2 antibody) binds to an antigen on the surface of a tumor cell. In this depiction, vIgD is a variant of the IgSF domain of CD80. As shown, the binding of the antibody to the surface of the tumor cell localizes vIgD on the tumor cell surface, where it can interact with one or more of its cognate binding partners expressed on the surface of adjacent immune cells (e.g., T cells) to activate receptor signaling. In the exemplary embodiment as shown, the variant IgSF domain (vIgD) is a variant of the IgSF domain of CD80. The binding of CD80 vIgD to the CD28 costimulatory receptor provides an agonistic or costimulatory signal. Figures 6A-6C show various exemplary configurations of a variant IgSF domain (vIgD) conjugated to an antibody (V-Mab). Figure 6A shows various configurations in which vIgD is linked directly or indirectly to the N-terminus and/or C-terminus of the light chain of the antibody. FIG. 6B shows various configurations in which vIgD is linked directly or indirectly to the N-terminus and/or C-terminus of the antibody heavy chain. FIG. 6C shows the resulting V-Mab configuration when the light chain of FIG. 6A and the heavy chain of FIG. 6B are co-expressed in cells. Figure 2 shows the results of a competitive binding assay for the binding of biotinylated recombinant CD28 Fc fusion protein (rCD28.Fc) to immobilized variant CD80 A91G ECD-Fc fusion molecules in the presence of unlabeled recombinant human PD-L1-his, human CTLA-4-his, or human-PD-L2-Fc fusion proteins. Figure 2 shows the results of a competitive binding assay for the binding of biotinylated recombinant human PD-L1-his monomer protein to immobilized variant CD80 A91G ECD-Fc fusion molecules in the presence of unlabeled recombinant human rCD28.Fc, human CTLA-4.Fc or human PD-L2.Fc. Figure 1 shows impedance results reflecting the cytotoxic killing activity of cells modified with anti-CD19 chimeric antigen receptor (CAR) alone or with exemplary transmembrane immunomodulatory TIP (CD80-TIP or ICOSL-TIP) or corresponding CD80 or ICOSL wild-type transmembrane proteins, followed by co-culture with target antigen-expressing cells. Impedance was evaluated using Acea Real-Time Cell Analyzer (RTCA) that measures impedance fluctuations in the culture medium of 96-well microelectronic plates (E-plates).

DETAILED DESCRIPTION Provided herein are immunomodulatory proteins that are or include variants of CD80 or mutants of CD80 and specific binding fragments thereof that have altered binding activity or affinity to at least one target ligand cognate binding partner (also referred to as counter-structure ligand protein). In some embodiments, the variant CD80 polypeptide contains one or more amino acid modifications (e.g., amino acid substitutions, deletions, or additions) compared to unmodified or wild-type CD80 polypeptide. In some embodiments, the variant CD80 polypeptide contains one or more amino acid modifications (e.g., substitutions) compared to unmodified or wild-type CD80 polypeptide. In some embodiments, the one or more amino acid modifications (e.g., substitutions) are in the IgSF domain (e.g., IgV) of unmodified or wild-type CD80 polypeptide. In some embodiments, the altered binding activity or affinity (e.g., improved or decreased binding activity or affinity) is to at least one of the cognate binding partner proteins CD28, PD-L1, or CTLA-4. In some embodiments, the variant CD80 polypeptide has an altered (e.g., improved or decreased) binding activity or affinity for one or more of CD28, PD-L1, or CTLA-4 compared to unmodified or wild-type CD80 that does not contain the one or more modifications. In some embodiments, the immunomodulatory protein is soluble. In some embodiments, the immunomodulatory protein is a transmembrane immunomodulatory protein that can be expressed on the surface of a cell. In some embodiments, also provided herein is one or more other immunomodulatory proteins that are conjugates or fusions containing a variant CD80 polypeptide provided herein and one or more other moieties or polypeptides.

In some embodiments, the variant CD80 polypeptide has improved binding affinity to one or more of CD28, PD-L1, or CTLA-4 compared to unmodified or wild-type CD80 that does not contain one or more modifications. In some embodiments, the variant CD80 polypeptide has reduced binding affinity to one or more of CD28, PD-L1, or CTLA-4 compared to unmodified or wild-type CD80 that does not contain one or more modifications. In some embodiments, the variant CD80 polypeptide has improved binding affinity to one or more of CD28, PD-L1, or CTLA-4 and reduced binding affinity to another one or more of CD28, PD-L1, or CTLA-4 compared to unmodified or wild-type CD80 that does not contain one or more modifications.

In some embodiments, also provided herein are one or more other immunomodulatory proteins that are conjugates or fusions containing a variant CD80 polypeptide provided herein and one or more other moieties or polypeptides. In some embodiments, the variant CD80 polypeptides and immunomodulatory proteins modulate an immunological immune response, e.g., enhance or reduce an immune response. In some embodiments, the variant CD80 polypeptides and immunomodulatory proteins provided herein can be used to treat a disease or condition associated with a dysregulated immune response.

In some embodiments, the variant CD80 polypeptides provided regulate T cell activation through interaction with costimulatory signaling molecules. In general, antigen-specific T cell activation generally requires two separate signals. The first signal is provided by interaction of the T cell receptor (TCR) with major histocompatibility complex (MHC)-associated antigens present on antigen-presenting cells (APCs). The second signal is costimulatory for TCR engagement and is required to avoid T cell apoptosis or anergy.

In some embodiments, under normal physiological conditions, T cell-mediated immune responses are initiated by antigen recognition by the T cell receptor (TCR) and are regulated by a balance of costimulatory and inhibitory signals (e.g., immune checkpoint proteins). The immune system relies on immune checkpoints to prevent autoimmunity (i.e., self-tolerance) and protect tissues from excessive damage during an immune response (e.g., during an attack against a pathogenic infection). However, in some cases, these immune-modulating proteins can become dysregulated in diseases and conditions, including tumors, as a mechanism to evade the immune system.

In some embodiments, among the known T cell costimulatory receptors is CD28, which is a T cell costimulatory receptor for the ligands B7-1 (CD80) and B7-2 (CD86), which are both present on APCs. These same ligands can also bind to the inhibitory T cell receptor CTLA4 (cytotoxic T lymphocyte-associated protein 4) with higher affinity than to CD28; binding to CTLA-4 acts to downregulate the immune response. Similarly, CD80 can bind to programmed cell death ligand 1 (PD-L1). PD-L1 is also a negative regulator of immune activation and can downregulate the immune response through interaction with the programmed cell death 1 (PD-1) receptor. Binding of CD80 to PD-L1 can block the interaction between PD-L1 and PD-1, thereby enhancing or potentiating the immune response. Thus, in some cases, the interaction of CD80 with CD28 and PD-L1 results in overlapping and complementary effects.

In some embodiments, CD28 and PD-L1 may play complementary roles in regulating immune responses. Enhancement or inhibition of the activity of these receptors has clinical significance for the treatment of inflammatory and autoimmune disorders, cancer, and viral infections. However, in some cases, therapies that intervene and alter the costimulatory effect of both receptors are constrained by spatial localization requirements as well as size limitations imposed by the localization of the immune synapse. In some aspects, existing therapeutics, including antibody drugs, may not simultaneously interact with multiple target proteins involved in regulating these interactions. In addition, in some cases, existing therapeutics may only have the ability to antagonize, but not stimulate, the immune response. Additionally, pharmacokinetic differences between drugs that independently target one or the other of these two receptors may create difficulties in adequately maintaining the desired blood levels of such drug combinations throughout the course of treatment.

In some embodiments, the variant CD28 polypeptides or immune modulating proteins provided modulate (e.g., enhance or reduce) immune activity induced with or associated with the costimulatory receptors CD28 or PD-L1 and, in some cases, CTLA-4. Thus, in some embodiments, the provided polypeptides overcome these limitations by providing variant CD80s that have independent binding affinities for both CD28 and PD-L1, and, in some cases, CTLA-4, thereby stimulating or antagonizing the complementary effects of costimulation by the receptors. Also provided are methods of making and using these CD80 variants.

All publications referred to herein (including patents, patent applications, scientific articles and databases) are incorporated herein by reference in their entirety for all purposes to the same extent as if each individual publication (including patents, patent applications, scientific articles and databases) was specifically and individually indicated to be incorporated by reference. To the extent that a definition set forth herein is contrary to or otherwise inconsistent with a definition set forth in a patent, application, published application and other publication incorporated herein by reference, the definition set forth herein takes precedence over the definition incorporated herein by reference.

The section headings used herein are for organizational purposes only and should not be construed as limiting the subject matter described.

I. Definitions
Unless otherwise defined, all technical terms, notes, and other technical and scientific terms or related terms used herein are intended to have the same meaning as commonly understood by one of ordinary skill in the art to which the claimed subject matter belongs. In some cases, terms having commonly understood meanings are defined herein for clarity and/or ready reference, and the inclusion of such definitions herein should not necessarily be construed as significantly different from those commonly understood in the art.

Terms used throughout this specification are defined as follows, unless otherwise limited in specific instances. As used in this specification and the appended claims, the singular forms "a," "an," and "the" include plural referents unless the context clearly dictates otherwise. Unless otherwise defined, all technical and scientific terms, acronyms, and abbreviations used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Unless otherwise indicated, abbreviations and symbols for chemical and biochemical names are in accordance with the IUPAC-IUB nomenclature system. Unless otherwise indicated, all numerical ranges include not only the values defining the range but also all integer values therebetween.

The term "affinity modified" when used in the context of immunoglobulin superfamily domains refers to a mammalian immunoglobulin superfamily (IgSF) domain having an amino acid sequence that has been altered to improve or decrease (relative to the corresponding wild-type parent or unmodified IgSF domain) the binding affinity or avidity for at least one of its cognate binding partners (or "counterstructures") compared to a parent wild-type or unmodified (i.e., affinity unmodified) IgSF control domain. In this context, affinity modified CD80 IgSF domains are included. In some embodiments, affinity modified IgSF domains can contain 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30 or more amino acid differences (e.g., amino acid substitutions) in a wild-type or unmodified IgSF domain. The improved or decreased binding affinity or avidity can be determined using well-known binding assays, such as flow cytometry. Larsen et al., American Journal of Transplantation, Vol 5: 443-453 (2005). See also Linsley et al., Immunity, 1: 7930801 (1994). The improved binding affinity or avidity of a protein to its cognate binding partner is at least 10% greater than the wild-type IgSF domain control, and in some embodiments, at least 20%, 30%, 40%, 50%, 100%, 200%, 300%, 500%, 1000%, 5000%, or 10000% greater than the wild-type IgSF domain control value. The reduction in the binding affinity or avidity of the protein for at least one of its cognate binding partners is 90% or less of the control but 10% or more of the wild-type IgSF domain control value, in some embodiments, 80%, 70%, 60%, 50%, 40%, 30%, or 20% or less of the wild-type IgSF domain control value but 10% or more. Affinity-modified proteins have a primary amino acid sequence altered by substitution, addition, or deletion of amino acid residues. The term "affinity-modified IgSF domain" should not be construed as imposing any conditions of any particular starting composition or method by which the affinity-modified IgSF domain was made. Thus, the affinity-modified IgSF domains of the present invention are not limited to any particular affinity modification process that converts a wild-type IgSF domain into an affinity-modified IgSF domain. Affinity-modified IgSF domain polypeptides can be generated, for example, starting from wild-type mammalian IgSF domain sequence information, then modeled in silico for binding to its cognate binding partner, and finally recombinantly or chemically synthesized to generate the subject affinity-modified IgSF domain compositions. As just another example, affinity-modified IgSF domains can be generated by site-directed mutagenesis of wild-type IgSF domains. Thus, affinity-modified IgSF domains represent products that are, but are not necessarily, produced by any given process. A variety of techniques may be used, including recombinant methods, chemical synthesis, or a combination thereof.

The term "allogeneic" as used herein means cells or tissues that are removed from one organism and then injected or adoptively transferred into a genetically distinct organism of the same species. In some embodiments of the invention, the species is murine or human.

The term "autologous" as used herein means cells or tissues that are removed from the same organism and then injected or adoptively transferred into said organism. Autologous cells or tissues can be modified, for example, by recombinant DNA techniques, such that they are no longer genetically identical to the native cells or native tissues removed from the organism. For example, native autologous T cells can be genetically modified by recombinant DNA techniques to become autologous modified cells that express transmembrane immunomodulatory proteins and/or chimeric antigen receptors (CARs), which in some cases involves modifying T cells or TILs (tumor infiltrating lymphocytes). The modified cells are then infused into the patient from which the native T cells were isolated. In some embodiments, the organism is a human or a mouse.

The terms "binding affinity" and "binding avidity" as used herein refer to the specific binding affinity and specific binding avidity, respectively, of a protein to its counter-structure under specific binding conditions. In biochemical kinetics, avidity refers to the cumulative strength of the affinities of multiple individual non-covalent interactions, such as between CD80 and its counter-structures PD-L1, CD28 and/or CTLA-4. Thus, avidity differs from affinity, which represents the strength of a single interaction. The improved or decreased binding affinity of an affinity-modified CD80 variant containing an IgSF domain to its counter-structure is determined relative to the binding affinity of an unmodified CD80 (e.g., an unmodified CD80 containing a native or wild-type IgSF domain (e.g., an IgV domain)). Methods for determining binding affinity or avidity are known in the art. See, e.g., Larsen et al., American Journal of Transplantation, Vol 5: 443-453 (2005). In some embodiments, the variant CD80 (e.g., a CD80 containing an affinity-modified IgSF domain) specifically binds to CD28, PD-L1 and/or CTLA-4 with a binding affinity that results in a mean fluorescence intensity (MFI) value that is at least 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, or 100% greater than an unmodified CD80 control, as measured by flow cytometry, in a binding assay, e.g., as described in Example 6.

The term "biological half-life" refers to the time it takes for a substance (e.g., an immune modulating polypeptide, including a variant CD80 polypeptide of the invention) to lose half of its pharmacological or physiological activity or concentration. Biological half-life can be affected by the elimination, excretion, degradation (e.g., enzymatic degradation) of the substance, or absorption and concentration in specific organs or tissues in the body. In some embodiments, biological half-life can be assessed by determining the time it takes for the plasma concentration of the substance to reach half of its steady-state level ("plasma half-life"). Conjugates that can be used to derivatize the polypeptides of the invention to extend their biological half-life are known in the art and include, but are not limited to, polyethylene glycol (PEG), hydroxyethyl starch (HES), XTEN (extended recombinant peptides; see WO 2013130683), human serum albumin (HSA), bovine serum albumin (BSA), lipids (acylated), and poly-Pro-Ala-Ser (PAS), polyglutamic acid (glutamylated).

The term "chimeric antigen receptor" or "CAR" as used herein refers to an artificial (i.e., man-made) transmembrane protein expressed on a mammalian cell, comprising at least an ectodomain, a transmembrane domain, and an endodomain. Optionally, the CAR protein comprises a "spacer" that covalently links the ectodomain to the transmembrane domain. The spacer is often a polypeptide that links the ectodomain to the transmembrane domain via a peptide bond. CARs are typically expressed on mammalian lymphocytes. In some embodiments, the CAR is expressed on a mammalian cell, such as a T cell or a tumor-infiltrating lymphocyte (TIL). A CAR expressed on a T cell is referred to herein as a "CAR T cell" or "CAR-T". In some embodiments, the CAR-T is a helper T cell, a cytotoxic T cell, a natural killer T cell, a memory T cell, a regulatory T cell, or a gamma delta T cell. When used clinically, for example in adoptive cell transfer, CAR-Ts with antigen-binding specificity for a patient's tumor are typically engineered to express on natural T cells obtained from the patient. The engineered T cells expressing the CAR are then infused back into the patient. Thus, the CAR-T is often an autologous CAR-T, although allogeneic CAR-Ts are also included within the scope of the present invention. The ectodomain of the CAR comprises an antigen-binding region (e.g., an antibody or an antigen-binding fragment thereof (e.g., scFv)) that specifically binds to a target antigen (e.g., a tumor-specific antigen) under physiological conditions. Upon specific binding, a series of biochemical events (i.e., signal transduction) results in the regulation of the immune activity of the CAR-T. Thus, for example, specific binding to its target antigen by the antigen-binding region of the CAR-T can lead to changes in the immune activity of T cell activity, as reflected by changes in cytotoxicity, proliferation, or cytokine production. In some embodiments, signal transduction upon CAR-T activation is achieved by the CD3 zeta chain ("CD3-z"), which is involved in signal transduction in natural mammalian T cells. CAR-Ts can further contain multiple signaling domains (e.g., CD28, 41BB, or OX40) that further modulate the immunoregulatory response of T cells. CD3-z contains a conserved motif known as an immunoreceptor tyrosine-based activation motif (ITAM) that is involved in T cell receptor signaling.

The term "total" or "total," when used in reference to cytokine production induced by the presence of two or more variant CD80 polypeptides in an in vitro assay, refers to the overall cytokine expression level regardless of cytokine production induced by individual variant CD80 polypeptides. In some embodiments, the cytokine assayed is IFN-γ, e.g., in an in vitro primary T cell assay as described in Example 7.

The term "cognate binding partner" (used interchangeably with "counterstructure"), with respect to a polypeptide (e.g., an IgSF domain of a variant CD80), refers to at least one molecule (typically a native mammalian protein) to which the referenced polypeptide specifically binds under specific binding conditions. In some aspects, a variant CD80 containing an affinity-modified IgSF domain specifically binds with improved or reduced affinity to a counterstructure of the corresponding native or wild-type CD80. A species of ligand that is recognized and specifically binds to its cognate receptor under specific binding conditions is an example of a counterstructure or cognate binding partner of that receptor. A "cell surface cognate binding partner" is a cognate binding partner expressed on the surface of a mammalian cell. A "cell surface molecular species" is a cognate binding partner of a ligand of the immune synapse (IS) that is expressed on or expressed by a cell (e.g., a mammalian cell) that forms the immune synapse (IS).

As used herein, "conjugate," "conjugation," or grammatical variations thereof, refers to the joining or linking of two or more compounds together, by any connecting or linking method known in the art, resulting in the formation of another compound. It can also refer to a compound produced by joining or linking two or more compounds together. For example, a variant CD80 polypeptide directly or indirectly linked to one or more chemical moieties or polypeptides is an exemplary conjugate. Such conjugates include fusion proteins, those produced by chemical conjugation, and those produced by any other method.

The term "competitive binding" as used herein means that a protein can specifically bind to at least two cognate binding partners, but the specific binding of one cognate binding partner inhibits (e.g., interferes with or prevents) the simultaneous binding of a second cognate binding partner. Thus, in some cases, a protein cannot simultaneously bind to two cognate binding partners. Generally, competitive binders contain the same or overlapping binding sites for specific binding, although this is not a requirement. In some embodiments, competitive binding causes a measurable (partial or complete) inhibition of specific binding of a protein to one of its cognate binding partners due to the specific binding of a second cognate binding partner. Various methods of quantifying competitive binding are known, such as ELISA (enzyme-linked immunosorbent assay).

The term "conservative amino acid substitution," as used herein, refers to an amino acid substitution in which one amino acid residue is replaced by another amino acid residue having a side chain R group with similar chemical properties (e.g., charge or hydrophobicity). Examples of groups of amino acids with side chains with similar chemical properties include: 1) aliphatic side chains: glycine, alanine, valine, leucine, and isoleucine; 2) aliphatic-hydroxyl side chains: serine and threonine; 3) amide-containing side chains: asparagine and glutamine; 4) aromatic side chains: phenylalanine, tyrosine, and tryptophan; 5) basic side chains: lysine, arginine, and histidine; 6) acidic side chains: aspartic acid and glutamic acid; and 7) sulfur-containing side chains: cysteine and methionine. Conservative amino acid substitution groups are: valine-leucine-isoleucine, phenylalanine-tyrosine, lysine-arginine, alanine-valine, glutamate-aspartate, and asparagine-glutamine.

The term "corresponding" with respect to a protein position, e.g., a statement that a nucleotide or amino acid position "corresponds" to a nucleotide or amino acid position in a disclosed sequence (e.g., as shown in the sequence listing), refers to a nucleotide or amino acid position that is identified by alignment with the disclosed sequence based on a structural sequence alignment or using a standard alignment algorithm (e.g., the GAP algorithm). For example, corresponding residues can be identified by alignment with a reference sequence having the sequence of wild-type CD80 shown in SEQ ID NO: 28 (ECD domain) or the sequence shown in SEQ ID NO: 152 or 372 (IgV domain) by the structural alignment methods described herein. By aligning the sequences, one skilled in the art can identify corresponding residues, e.g., using conserved and identical amino acid residues as criteria.

The terms "reduce" or "attenuate" or "inhibit" as used herein mean a statistically significant amount of reduction. The reduction can be at least a 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, or 100% reduction.

The term "derivative" or "derivatized" refers to the modification of a protein by covalently attaching the protein directly or indirectly to a composition to alter properties such as biological half-life, bioavailability, immunogenicity, solubility, toxicity, potency, or efficacy while retaining or enhancing its therapeutic benefit. Derivatives of the immune modulating polypeptides of the invention are within the scope of the invention and can be produced, for example, by glycosylation, pegylation, lipidation, or Fc fusion.

As used herein, a domain (typically a sequence of 3 or more, generally 5 or 7 or more amino acids, e.g., 10-200 amino acid residues) refers to a portion of a molecule (e.g., a protein or coding nucleic acid) that is structurally and/or functionally distinct from and identifiable with other portions of the molecule. For example, a domain includes a portion of a polypeptide chain that can form an independently folded structure within a protein that is composed of one or more structural motifs and/or is recognized by a functional activity, such as binding activity. A protein can have one or more distinct domains. For example, a domain can be identified, defined, or distinguished by primary sequence or structural homology to related family members, e.g., homology to a motif. In another example, a domain can be distinguished by its function (e.g., ability to interact with a biomolecule, such as a cognate binding partner). A domain can exhibit a biological function or activity independently, such that a domain can perform an activity (e.g., binding) independently or fused to another molecule. A domain can be a linear or non-linear amino acid sequence. Many polypeptides contain multiple domains. Such domains are known and can be identified by one of ordinary skill in the art. For purposes of illustration herein, definitions are provided, but it is understood that it is well within the skill of the art to recognize a particular domain by name. If necessary, appropriate software can be employed to identify the domains.

The term "ectodomain," as used herein, refers to a region of a membrane protein (e.g., a transmembrane protein) that is outside the vesicle membrane. Ectodomains often contain a binding domain that specifically binds to a ligand or cell surface receptor, e.g., via a binding domain that specifically binds to the ligand or cell surface receptor. The ectodomain of a cellular transmembrane protein is alternatively referred to as the extracellular domain.

The term "effective amount" or "therapeutically effective amount" refers to an amount and/or concentration of a therapeutic composition (including a protein or cell composition) of the present invention that, when administered ex vivo (by contact with cells from a patient) or in vivo (by administration to a patient) either alone (i.e., as a monotherapy) or in combination with an additional therapeutic agent, results in a statistically significant reduction in disease progression, e.g., by ameliorating or eliminating the symptoms and/or pathogenesis of the disease. An effective amount can be an amount that alleviates, reduces, or relieves at least one symptom or biological response or effect associated with a disease or disorder, prevents the progression of a disease or disorder, or improves the physical function of a patient. In the case of cell therapy, an effective amount is an effective dose or number of cells administered to a patient by adoptive cell therapy. In some embodiments, the patient is a mammalian patient, e.g., a non-human primate or a human patient.

The term "endodomain" as used herein refers to a region found in some membrane proteins (e.g., transmembrane proteins) that extends into the interior space defined by the cell surface membrane. In mammalian cells, the endodomain is the cytoplasmic region of the membrane protein. Within the cell, the endodomain can interact with intracellular components and play a role in signal transduction, and thus, in some cases, can be an intracellular signaling domain. The endodomain of a cellular transmembrane protein is alternatively referred to as a cytoplasmic domain, which in some cases can be a cytoplasmic signaling domain.

The term "enhanced" or "increased" as used herein in the context of enhancing immune activity of a mammalian lymphocyte means enhancing one or more activities of the lymphocyte. The enhanced activity can be an enhancement (e.g., a statistically significant amount) of one or more of cell survival, cell proliferation, cytokine production, or T-cell cytotoxicity. In some embodiments, reference to enhanced immune activity means increasing (e.g., by a statistically significant amount) interferon gamma (IFNγ) production. In some embodiments, immune activity can be assessed in a mixed lymphocyte reaction (MLR) assay. Methods for performing MLR assays are known in the art. Wang et al., Cancer Immunol Res. 2014 Sep: 2(9):846-56. Other methods for assessing lymphocyte activity are known in the art, including any of the assays as described herein. In some embodiments, the enhancement can be an increase (improvement) of at least 10%, 20%, 30%, 40%, 50%, 75%, 100%, 200%, 300%, 400%, or 500% greater than a non-zero control value.

The term "modified cell" as used herein refers to a mammalian cell that has been genetically modified by human intervention (e.g., recombinant DNA or viral transduction). In some embodiments, the cell is an immune cell, such as a lymphocyte (e.g., T cell, B cell, NK cell) or an antigen-presenting cell (e.g., dendritic cell). The cell may be a primary cell from a patient or may be a cell line. In some embodiments, the modified cell of the invention comprises a variant CD80 polypeptide of the invention that has been modified to modulate the immune activity of a T cell expressing CD28, PD-L1, and/or CTLA-4 to which the variant CD80 specifically binds, or an APC expressing PD-L1. In some embodiments, the variant CD80 is a transmembrane immunomodulatory protein (hereinafter referred to as "TIP") that contains an extracellular domain or a portion thereof that contains an IgV domain linked to a transmembrane domain (e.g., a CD80 transmembrane domain), and optionally contains an intracellular signaling domain. In some cases, the TIP is configured as a chimeric receptor that contains a heterologous cytoplasmic signaling domain or endodomain. In some embodiments, the modified cells are capable of expressing and secreting an immunomodulatory protein as described herein. Some modified cells provided further contain a modified T cell receptor (TCR) or a chimeric antigen receptor (CAR).

The term "modified T cell," as used herein, refers to a T cell (e.g., a helper T cell, a cytotoxic T cell (or cytotoxic T lymphocyte or CTL), a natural killer T cell, a regulatory T cell, a memory T cell, or a gamma delta T cell) that has been genetically modified (altered) by human intervention (e.g., recombinant DNA or viral transduction). The modified T cell comprises a variant CD80 transmembrane immunomodulatory protein (TIP) of the invention expressed on the T cell, which has been modified to modulate the immune activity of the modified T cell itself or of a mammalian cell to which the variant CD80 expressed on the T cell specifically binds.

The term "modified T cell receptor" or "modified TCR" refers to a T cell receptor (TCR) that has been modified to specifically bind with a desired affinity to a major histocompatibility complex (MHC)/peptide target antigen that has been selected, cloned, and/or subsequently introduced into a population of T cells, which are often used for adoptive immunotherapy. In contrast to modified TCRs, CARs are modified to bind target antigens in an MHC-dependent manner.

The term "expressed on," as used herein, is used in reference to a protein expressed on the surface of a cell (e.g., a mammalian cell). Thus, the protein is expressed as a membrane protein. In some embodiments, the expressed protein is a transmembrane protein. In some embodiments, the protein is conjugated to a small molecule moiety (e.g., a drug or a detectable label). A protein expressed on the surface of a cell can include a cell surface protein (e.g., a cell surface receptor) expressed on a mammalian cell.

The term "half-life extending moiety" refers to a portion of a polypeptide fusion or chemical conjugate that extends the half-life of a protein circulating in mammalian serum compared to the half-life of a protein not so conjugated to the moiety. In some embodiments, the half-life is extended by more than 1.2-fold, 1.5-fold, 2.0-fold, 3.0-fold, 4.0-fold, 5.0-fold, or 6.0-fold, or by more than about 1.2-fold, about 1.5-fold, about 2.0-fold, about 3.0-fold, about 4.0-fold, about 5.0-fold, or about 6.0-fold. In some embodiments, the half-life is extended by more than 6 hours, more than 12 hours, more than 24 hours, more than 48 hours, more than 72 hours, more than 96 hours, or more than 1 week after in vivo administration compared to a protein without the half-life extending moiety. Half-life refers to the time it takes for a protein to lose half of its concentration, amount, or activity. Half-life can be determined, for example, by using an ELISA assay or an activity assay. Exemplary half-life extending moieties include Fc domains, multimerization domains, polyethylene glycol (PEG), hydroxyethyl starch (HES), XTEN (extended recombinant peptides; see WO 2013130683), human serum albumin (HSA), bovine serum albumin (BSA), lipids (acylation), and poly-Pro-Ala-Ser (PAS), and polyglutamic acid (glutamylation).

The term "immune synapse," as used herein, refers to the interface between a mammalian cell (e.g., an antigen-presenting cell or a tumor cell) that expresses MHC I (major histocompatibility complex) or MHC II, and a mammalian lymphocyte (e.g., an effector T cell or a natural killer (NK) cell).

The Fc (fragment crystallizable) region or domain of an immunoglobulin molecule (also called Fc polypeptide) corresponds primarily to the constant region of the immunoglobulin heavy chain and is involved in various functions, including the effector functions of an antibody. The Fc domain contains part or all of the hinge domain and the CH2 and CH3 domains of an immunoglobulin molecule. The Fc domain can form a dimer of two polypeptide chains connected by one or more disulfide bonds. In some embodiments, the Fc is a variant Fc with reduced activity (e.g., 30%, 40%, 50%, 60%, 70%, 80%, 90% or greater) in promoting effector function. In some embodiments, references to amino acid substitutions in the Fc region are by EU numbering system, unless otherwise noted with reference to a specific SEQ ID NO. EU numbering is known and follows the EU index as reported in the recently updated IMGT Scientific Chart (IMGT®, i.e., international ImMunoGeneTics information system®, http://www.imgt.org/IMGTScientificChart/Numbering/Hu_IGHGnber.html (created May 17, 2001, last updated January 10, 2013) and Kabat, E.A. et al. Sequences of Proteins of Immunological interest. 5th ed. US Department of Health and Human Services, NIH publication No. 91-3242 (1991).

Immunoglobulin Fc fusions ("Fc fusions"), e.g., immunomodulatory Fc fusion proteins, are molecules that comprise one or more polypeptides (or one or more small molecules) operatively linked to the Fc region of an immunoglobulin. Fc fusions can comprise, for example, an Fc region of an antibody (to facilitate pharmacokinetics) and a variant CD80 polypeptide. The immunoglobulin Fc region can be indirectly or directly linked to one or more variant CD80 polypeptides or small molecules (fusion partners). A variety of linkers are known in the art and can optionally be used to link the Fc to the fusion partner to generate the Fc fusion. Fc fusions of the same species can be dimerized to form Fc fusion homodimers, or non-identical species can be used to form Fc fusion heterodimers. In some embodiments, the Fc is a mammalian Fc, e.g., a murine or human Fc.

The term "host cell" refers to a cell that can be used to express a protein encoded by a recombinant expression vector. A host cell can be a prokaryote, such as E. coli, or a host cell can be a eukaryote, such as a unicellular eukaryote (e.g., yeast or other fungi), a plant cell (e.g., tobacco or tomato plant cell), an animal cell (e.g., a human cell, a monkey cell, a hamster cell, a rat cell, a mouse cell, or an insect cell) or a hybridoma. Examples of host cells include Chinese hamster ovary (CHO) cells or their derivatives, such as Veggie CHO and related cell lines that grow in serum-free medium or the CHO line DX-B11, which is DHFR-deficient. In some embodiments, the host cell may be a mammalian cell (e.g., a human cell, a monkey cell, a hamster cell, a rat cell, a mouse cell, or an insect cell).

The term "immunoglobulin" (abbreviated as "Ig"), as used herein, refers to mammalian immunoglobulin proteins, including any of the five human classes of antibodies: IgA (including subclasses IgA1 and IgA2), IgD, IgE, IgG (including subclasses IgG1, IgG2, IgG3, and IgG4), and IgM. The term also includes immunoglobulins that are less than full length, whether fully or partially synthetic (e.g., recombinantly or chemically synthesized) or naturally produced, such as antigen-binding fragments (Fab), variable fragments (Fv) containing _VH and _VL , single-chain variable fragments (scFv) containing _VH and _VL linked together in one chain, and other antibody V region fragments (Fab', F(ab) ₂ , F(ab') ₂ , dsFv diabodies, Fc, and Fd polypeptide fragments). Homo- and hetero-bispecific bispecific antibodies are included within the scope of the term.

The term "immunoglobulin superfamily" or "IgSF" as used herein refers to a group of cell surface and soluble proteins involved in cellular recognition, binding, or adhesion processes. Molecules are categorized as members of this superfamily based on structural features they share with immunoglobulins (i.e., antibodies); they all possess domains known as immunoglobulin domains or folds. Members of IgSF include cell surface antigen receptors, co-receptors and co-stimulatory molecules of the immune system, molecules involved in antigen presentation to lymphocytes, cell adhesion molecules, certain cytokine receptors, and intracellular muscle proteins. These are usually associated with a role in the immune system. Proteins in the immune synapse are often members of IgSF. IgSF can also be divided into "subfamilies" based on shared properties such as function. Such subfamilies typically consist of 4-30 IgSF members.

The term "IgSF domain" or "immunoglobulin domain" or "Ig domain" as used herein refers to a structural domain of the IgSF protein. Ig domains are named after immunoglobulin molecules. They contain about 70-110 amino acids and are categorized according to their size and function. Ig domains possess a characteristic Ig fold, with a sandwich-like structure formed by two sheets of antiparallel β-strands. Interactions between hydrophobic amino acids on the interior of the sandwich and highly conserved disulfide bonds formed between cysteine residues in the B and F strands stabilize the Ig fold. One end of the Ig domain has a portion called the complementarity determining region, which is important for the specificity of antibodies for their ligands. Ig-like domains can be classified (into classes) as IgV, IgC1, IgC2, or IgI. Most Ig domains are either variable (IgV) or constant (IgC) domains. IgV domains, with nine β-strands, are generally longer than IgC domains, with seven β-strands. The Ig domains of some members of the IgSF resemble IgV domains in amino acid sequence, yet are similar in size to IgC domains. These are called IgC2 domains, while the canonical IgC domain is called IgC1 domain. T cell receptor (TCR) chains contain two Ig domains in their extracellular portions, one IgV domain at the N-terminus and one IgC1 domain adjacent to the cell membrane. CD80 contains two Ig domains: IgV and IgC.

The term "IgSF species" as used herein refers to a population of IgSF member proteins having identical or substantially identical primary amino acid sequences. Each mammalian immunoglobulin superfamily (IgSF) member defines a unique identity for all IgSF species belonging to that IgSF member. Thus, each IgSF family member is unique compared to other IgSF family members, and thus each species of a particular IgSF family member is unique compared to another IgSF family member species. Nevertheless, differences between molecules of the same IgSF species may arise due to differences in post-translational modifications such as glycosylation, phosphorylation, ubiquitination, nitrosylation, methylation, acetylation, and lipidation. Furthermore, small sequence differences within a single IgSF species due to genetic polymorphisms, as well as wild-type truncated forms of an IgSF species due to, for example, proteolytic cleavage, also constitute differences between other forms within a single IgSF species. A "cell surface IgSF species" is an IgSF species that is expressed on the surface of a cell, typically a mammalian cell.

The term "immune activity," as used herein in the context of mammalian lymphocytes such as T cells, refers to one or more of cell survival, cell proliferation, cytokine production (e.g., interferon-γ), or T-cytotoxic activity. In some cases, immune activity can refer to expression of cytokines such as chemokines or interleukins. Assays for determining enhanced or suppressed immune activity include the MLR (mixed lymphocyte reaction) assay, which measures interferon-γ cytokine levels in culture supernatants (Wang et al., Cancer Immunol Res. 2014 Sep: 2(9):846-56), the SEB (staphylococcal enterotoxin B) T cell stimulation assay (Wang et al., Cancer Immunol Res. 2014 Sep: 2(9):846-56), and the anti-CD3 T cell stimulation assay (Li and Kurlander, J Transl Med. 2010: 8: 104). Since T cell activation is associated with the secretion of IFN-γ cytokines, detection of IFN-γ levels in culture supernatants from these in vitro human T cell assays can be assayed using commercially available ELISA kits (Wu et al, Immunol Lett 2008 Apr 15; 117(1): 57-62). Induction of an immune response results in enhanced immune activity compared to resting lymphocytes. An immune modulating protein as provided herein (e.g., a variant CD80 polypeptide containing an affinity-modified IgSF domain) can, in some embodiments, increase, or in alternative embodiments decrease, IFN-γ (interferon-γ) expression in primary T cell assays compared to wild-type IgSF members or IgSF domain controls. One skilled in the art will recognize that the format of the primary T cell assay used to determine increased IFN-γ expression will differ from the format employed to assay for decreased IFN-γ expression. In assaying for the ability of the immunomodulatory proteins or affinity-modified IgSF domains of the invention to reduce IFN-γ expression in primary T cell assays, a mixed lymphocyte reaction (MLR) assay can be used as described in Example 6. Conveniently, a soluble form of the affinity-modified IgSF domain of the invention can be employed to determine its ability to reduce IFN-γ expression by antagonizing it in an MLR, also as described in Example 6. Alternatively, in assaying for the ability of the immunomodulatory proteins or affinity-modified IgSF domains of the invention to increase IFN-γ expression in primary T cell assays, a co-fixation assay can be used. In the co-fixation assay, a T cell receptor signal (provided in some embodiments by an anti-CD3 antibody) is used in combination with a co-fixed affinity-modified IgSF domain, such as a variant CD80, to determine the ability to increase IFN-γ expression compared to a wild-type IgSF domain control. Methods for assaying the immune activity of modified cells, including assessing the activity of variant CD80 transmembrane immunomodulatory proteins, are known in the art and include, but are not limited to, the ability to expand T cells after antigen stimulation, the ability to sustain T cell proliferation in the absence of restimulation, and anti-cancer activity in appropriate animal models. Assays also include assays to assess cytotoxicity, including standard ⁵¹ Cr release assays (see, e.g., Milone et al., (2009) Molecular Therapy 17: 1453-1464) or flow-based cytotoxicity assays, or impedance-based cytotoxicity assays (Peper et al. (2014) Journal of Immunological Methods, 405:192-198).

An "immunomodulating polypeptide" is a polypeptide that modulates immune activity. "Modulating" an immune response means either enhancing or suppressing immune activity. An immunomodulating polypeptide can be a single polypeptide chain or a multimer (dimer or higher order multimer) of at least two polypeptide chains covalently linked to each other (e.g., by interchain disulfide bonds). Thus, monomeric, dimeric, and higher order multimeric polypeptides are within the scope of the defined term. Multimeric polypeptides can be homomultimers (of the same polypeptide chain) or heteromultimers (of different polypeptide chains). An immunomodulating polypeptide can include a variant CD80.

The term "increase" as used herein means to increase by a statistically significant amount. The increase can be at least 5%, 10%, 20%, 30%, 40%, 50%, 75%, 100%, or greater than a non-zero control value.

An "isoform" of CD80 is one of multiple naturally occurring CD80 polypeptides that differ in amino acid sequence. Isoforms can be the products of splice variants of an RNA transcript expressed by a single gene, or the expression products of highly similar but distinct genes that generate functionally similar proteins, such as may result from gene duplication. As used herein, the term "isoform" of CD80 also refers to the products of different alleles of the CD80 gene.

The term "lymphocyte" as used herein means any of three subtypes of white blood cells of the mammalian immune system. These include natural killer cells (NK cells) (which function in cell-mediated cytotoxic innate immunity), T cells (involved in cell-mediated cytotoxic adaptive immunity), and B cells (involved in humoral antibody-mediated adaptive immunity). T cells include helper T cells, cytotoxic T cells, natural killer T cells, memory T cells, regulatory T cells, or gamma delta T cells. Also included within the definition of lymphocytes are innate lymphoid cells (ILCs).

The term "mammal" or "patient" specifically includes reference to at least one of a human, chimpanzee, rhesus monkey, cynomolgus monkey, dog, cat, mouse, or rat.

The term "membrane protein" as used herein means a protein that directly or indirectly attaches (binds) to a lipid bilayer under physiological conditions. The lipid bilayer that forms the membrane can be a biological membrane, such as a cell membrane of a eukaryotic organism (e.g., a mammalian organism) or an artificial (i.e., man-made) membrane, such as that found on a liposome. The binding of the membrane protein to the lipid bilayer can be by covalent bonds or by non-covalent interactions (e.g., hydrophobic or electrostatic interactions). The membrane protein can be an integral membrane protein or a peripheral membrane protein. A membrane protein that is a peripheral membrane protein is bound to the lipid bilayer by non-covalent interactions or is bound to an integral membrane protein by non-covalent interactions. The peripheral membrane protein forms a temporary binding to the lipid bilayer so that the peripheral membrane protein can interact with and/or dissociate from the lipid bilayer under conditions in a physiological range in a mammal. In contrast to peripheral membrane proteins, integral membrane proteins form a substantially permanent attachment to the lipid bilayer of the membrane such that the integral membrane protein does not dissociate from the attachment to the lipid bilayer under conditions in the physiological range in mammals. A membrane protein can form an attachment to the membrane by one layer of the lipid bilayer (monotopic) or by both layers of the membrane (polytopic). An integral membrane protein that interacts with only one lipid bilayer is an "integral monotopic protein." An integral membrane protein that interacts with both lipid bilayers is an "integral polytopic protein." Alternatively, they are referred to herein as "transmembrane proteins."

The term "modulation" or "modulating," as used herein in the context of an immune response (e.g., a mammalian immune response), refers to any change (e.g., enhancement or reduction) in an existing or potential immune response that occurs as a result of administration of an immunomodulatory polypeptide comprising a variant CD80 of the invention or as a result of administration of a modified cell expressing an immunomodulatory protein of the invention (e.g., a variant CD80 transmembrane immunomodulatory protein). Thus, modulation refers to a change (e.g., enhancement or reduction) in an immune response compared to an immune response that occurs or exists in the absence of administration of an immunomodulatory protein comprising a variant CD80. Such modulation includes any induction, activation, suppression, or change in a degree or extent of immune activity of an immune cell. Immune cells include B cells, T cells, NK (natural killer) cells, NK T cells, professional antigen presenting cells (APCs), and non-professional antigen presenting cells, as well as inflammatory cells (neutrophils, macrophages, monocytes, eosinophils, and basophils). Modulation includes any change made to an existing, developing, or potential immune response, or to the ability to induce, modulate, affect, or respond to an immune response. Modulation includes any change in the expression and/or function of genes, proteins, and/or other molecules in immune cells as part of an immune response. Modulation of immune responses or modulation of immune activity includes, for example, the elimination, deletion, or sequestration of immune cells; induction or generation of immune cells that can modulate the functional capabilities of other cells, such as autoreactive lymphocytes, antigen-presenting cells, or inflammatory cells; induction of an unresponsive state in immune cells (i.e., anergy); enhancing or suppressing the activity or function of immune cells (including, but not limited to, changing the pattern of proteins expressed by these cells). Examples include changes in the production and/or secretion of specific classes of molecules, such as cytokines, chemokines, growth factors, transcription factors, kinases, costimulatory molecules, or other cell surface receptors, or any combination of these regulatory events. Modulation can be assessed, for example, by changes in IFN-γ (interferon gamma) expression compared to wild-type or unmodified CD80 controls in primary T cell assays (see Zhao and Ji, Exp Cell Res. 2016 Jan1; 340(1) 132-138). Modulation can be assessed, for example, by changes in immune activity of the modified cells, for example changes in cytotoxic activity of the modified cells or changes in cytokine secretion of the modified cells, compared to cells modified with wild-type CD80 transmembrane protein.

The term "multimerization domain" refers to an amino acid sequence that promotes stable interaction of a polypeptide molecule with one or more additional polypeptide molecules, each of which contains a complementary multimerization domain, which can be the same or a different multimerization domain that forms a stable multimer with the first domain. Generally, the polypeptide is directly or indirectly connected to the multimerization domain. Exemplary multimerization domains include immunoglobulin sequences or portions thereof, leucine zippers, hydrophobic regions, hydrophilic regions, and compatible protein-protein interaction domains. The multimerization domain can be, for example, an immunoglobulin constant region or domain, such as an Fc domain or portion thereof, from IgG (including IgG1, IgG2, IgG3, or IgG4 subtypes), IgA, IgE, IgD, and IgM, and modified forms thereof.

The terms "nucleic acid" and "polynucleotide" are used interchangeably and refer to a polymer of nucleic acid residues (e.g., deoxyribonucleotides or ribonucleotides) in either single-stranded or double-stranded form. Unless otherwise limited, the terms encompass nucleic acids that contain analogs of known natural nucleotides and have similar binding properties thereto, and that are metabolized in a manner similar to naturally occurring nucleotides. Unless otherwise specified, a particular nucleic acid sequence also implicitly encompasses not only the sequence explicitly shown (the "reference sequence"), but also conservatively modified variants thereof (e.g., degenerate codon substitutions) and complementary nucleotide sequences. Specifically, degenerate codon substitutions may be achieved by generating sequences in which the third position of one or more selected (or all) codons is substituted with mixed base and/or deoxyinosine residues. The term nucleic acid or polynucleotide encompasses cDNA or mRNA encoded by a gene.

The term "molecular species" as used herein means a population of proteins with identical or substantially identical primary amino acid sequences. Each mammalian immunoglobulin superfamily (IgSF) member defines a collection of identical or substantially identical molecular species. Thus, for example, human CD80 is an IgSF member and each human CD80 molecule is a molecular species of CD80. Differences between molecules of the same molecular species can occur due to differences in post-translational modifications such as glycosylation, phosphorylation, ubiquitination, nitrosylation, methylation, acetylation, and lipidation. Furthermore, small sequence differences within a single molecular species due to genetic polymorphisms, as well as wild-type truncated forms of a single molecular species due to, for example, proteolytic cleavage, also constitute other forms of differences within a single molecular species. A "cell surface molecular species" is a molecular species that is expressed on the surface of a mammalian cell. Two or more different protein species, each present in only one or only the other (but not both) of two mammalian cells forming an IS, are said to be in "cis" or in a "cis configuration" with respect to each other. Two distinct protein species, the first of which is present only in the first of two mammalian cells that form an IS and the second of which is present only in the second of two mammalian cells that form an IS, are said to be in "trans" or in a "trans configuration." Two distinct protein species, each of which is present in both of two mammalian cells that form an IS, are in both cis and trans configurations on those cells.

The term "non-competitive binding" as used herein refers to the ability of a protein to specifically bind at least two cognate binding partners simultaneously. Thus, although a protein can simultaneously bind at least two different cognate binding partners, the binding interactions need not be of the same duration, so in some cases the protein is specifically bound to only one of the cognate binding partners. In some embodiments, the binding occurs under specific binding conditions. In some embodiments, the simultaneous binding is such that the binding of one cognate binding partner does not substantially inhibit the simultaneous binding to a second cognate binding partner. In some embodiments, non-competitive binding means that the binding of a second cognate binding partner to its binding site on the protein does not displace the binding of a first cognate binding partner to its binding site on the protein. Methods for assessing non-competitive binding are well known in the art, such as those described in Perez de La Lastra et al., Immunology, 1999 Apr: 96(4): 663-670. In some cases, in a non-competitive interaction, a first cognate binding partner specifically binds at an interaction site that does not overlap with the interaction site of a second cognate binding partner, such that the binding of the second cognate binding partner does not directly interfere with the binding of the first cognate binding partner. Thus, any effect of the binding of the second cognate binding partner on the binding of the cognate binding partner is through a mechanism other than direct interference with the binding of the first cognate binding partner. For example, in the context of an enzyme-substrate interaction, a non-competitive inhibitor binds to a site other than the active site of the enzyme. Non-competitive binding encompasses non-competitive binding interactions in which a second cognate binding partner specifically binds at an interaction site that does not overlap with the binding of the first cognate binding partner, but binds to the second interaction site only when the first interaction site is occupied by the first cognate binding partner.

The term "pharmaceutical composition" refers to a composition suitable for pharmaceutical use in a mammalian subject, often a human. A pharmaceutical composition typically comprises an effective amount of an active agent (e.g., an immunomodulatory polypeptide comprising a variant CD80 or a modified cell expressing a variant CD80 transmembrane immunomodulatory protein) and a carrier, excipient, or diluent. The carrier, excipient, or diluent is typically a pharma- ceutically acceptable carrier, excipient, or diluent, respectively.

The terms "polypeptide" and "protein" are used interchangeably herein and refer to a molecular chain of two or more amino acids linked via peptide bonds. The term does not refer to a specific length of the product. Thus, "peptide" and "oligopeptide" are included within the definition of a polypeptide. The term includes post-translational modifications of the polypeptide, such as glycosylation, acetylation, phosphorylation, and the like. The term also includes molecules that are one or more amino acid analogs or non-standard or unnatural amino acids, which can be synthesized or recombinantly expressed using known protein modification techniques. In addition, proteins may be derivatized.

The term "primary T cell assay" as used herein refers to an in vitro assay for measuring interferon-gamma ("IFN-γ") expression. A variety of such primary T cell assays are known in the art, such as the assay described in Example 6. In a preferred embodiment, the assay used is an anti-CD3 co-fixation assay. In this assay, primary T cells are stimulated with fixed anti-CD3, with or without additional recombinant proteins. At a time point (usually 24-72 hours), culture supernatants are collected. In another embodiment, the assay used is a mixed lymphocyte reaction (MLR). In this assay, primary T cells are stimulated with allogeneic APCs. At a time point (usually 24-72 hours), culture supernatants are collected. Human IFN-γ levels in the culture supernatants are measured by standard ELISA techniques. Commercial kits are available from suppliers, and the assay is performed according to the manufacturer's recommendations.

The term "purified," as applied to nucleic acids (e.g., nucleic acids encoding the immunomodulatory proteins of the invention), generally refers to a nucleic acid or polypeptide that is substantially free of other components as determined by analytical techniques well known in the art (e.g., a purified polypeptide or polynucleotide forms a discrete band in an electrophoretic gel, a chromatographic eluate, and/or a medium subjected to density gradient centrifugation). For example, a nucleic acid or polypeptide that gives rise to essentially one band in an electrophoretic gel is "purified." A purified nucleic acid or protein of the invention is at least about 50% pure, typically at least about 75%, 80%, 85%, 90%, 95%, 96%, 99% or more pure (e.g., on a weight percent or molar basis).

The term "recombinant" indicates that a substance (e.g., a nucleic acid or polypeptide) has been artificially (i.e., non-naturally) altered by human intervention. The alteration can be performed on a substance in or removed from its natural environment or state. For example, a "recombinant nucleic acid" is one that is produced by recombining a nucleic acid, e.g., during cloning, affinity engineering, DNA shuffling or other well-known molecular biological procedures. A "recombinant DNA molecule" is composed of segments of DNA joined together by such molecular biological techniques. The term "recombinant protein" or "recombinant polypeptide" as used herein refers to a protein molecule expressed using a recombinant DNA molecule. A "recombinant host cell" is a cell that contains and/or expresses a recombinant nucleic acid or is otherwise modified by genetic engineering (e.g., by introducing into the cell a nucleic acid molecule encoding a recombinant protein, e.g., a transmembrane immunomodulatory protein provided herein). Transcriptional control signals in eukaryotes include "promoter" and "enhancer" elements. Promoters and enhancers consist of short arrays of DNA sequences that specifically interact with cellular proteins involved in transcription. Promoter and enhancer elements have been isolated from a variety of eukaryotic sources, including genes in yeast, insect and mammalian cells, and viruses (analogous control elements, i.e., promoters, are also found in prokaryotes). The selection of a particular promoter and enhancer depends on what cell type is to be used to express the protein of interest. The terms "in functional combination," "in functional order," and "operably linked," as used herein, refer to the linking of nucleic acid sequences in a manner or orientation that produces a nucleic acid molecule capable of directing the transcription of a given gene and/or the synthesis of a desired protein molecule.

The term "recombinant expression vector" as used herein refers to a DNA molecule that contains a desired coding sequence and appropriate nucleic acid sequences necessary for the expression of the operably linked coding sequence in a particular host cell. Nucleic acid sequences necessary for expression in prokaryotes include a promoter, optionally an operator sequence, a ribosome binding site, and possibly other sequences. Eukaryotic cells are known to utilize promoters, enhancers, and termination and polyadenylation signals. Optionally, a secretory signal peptide sequence can also be encoded by the recombinant expression vector operably linked to the coding sequence of the recombinant protein (e.g., recombinant fusion protein) so that the expressed fusion protein can be secreted by the recombinant host cell for easier isolation of the fusion protein from the cell. The term includes vectors as autonomously replicating nucleic acid structures as well as vectors that are integrated into the genome of a host cell into which they are introduced. Such vectors include viral vectors, such as lentiviral vectors.

The term "selectivity" refers to the preference of a protein or polypeptide of interest for specific binding to one substrate (e.g., one cognate binding partner) compared to the specific binding of the protein of interest to another substrate (e.g., a different cognate binding partner). Selectivity can be reflected as the ratio of the binding activity (e.g., binding affinity) (e.g., _Kd1 ) of a protein of interest to a first substrate (e.g., a first cognate binding partner) to the binding activity (e.g., binding affinity) (e.g., _Kd2 ) of the same protein of interest to a second cognate binding partner.

The term "sequence identity" as used herein refers to sequence identity at the nucleotide or amino acid level between genes or proteins, respectively. "Sequence identity" is a measure of identity between proteins at the amino acid level and between nucleic acids at the nucleotide level. Sequence identity of proteins can be determined by comparing the amino acid sequence at a given position in each sequence when the sequences are aligned. Similarly, sequence identity of nucleic acids can be determined by comparing the nucleotide sequence at a given position in each sequence when the sequences are aligned. Methods for alignment of sequences for comparison are well known in the art, such methods include GAP, BESTFIT, BLAST, FASTA, and TFASTA. The BLAST algorithm calculates percent sequence identity and performs a statistical analysis of the similarity between two sequences. Software for performing BLAST analysis is publicly available through the website of the National Center for Biotechnology Information (NCBI).

The term "soluble" as used herein with respect to a protein means that the protein is not a membrane protein. Generally, a soluble protein contains only the extracellular domain or a portion thereof of an IgSF family member receptor that contains an IgSF domain or a specific binding fragment thereof, but does not contain a transmembrane domain. In some cases, the solubility of the protein can be improved by linking or attaching, directly or indirectly via a linker, to an Fc domain, which in some cases can also improve the stability and/or half-life of the protein. In some aspects, the soluble protein is an Fc fusion protein.

The term "species" as used herein with respect to a polypeptide or nucleic acid refers to a population of molecules having identical or substantially identical sequences. Differences between polypeptides of the same species can arise due to differences in post-translational modifications such as glycosylation, phosphorylation, ubiquitination, nitrosylation, methylation, acetylation, and lipidation. Slightly truncated polypeptide sequences that differ from the full-length species by (or code for) only one, two, or three amino acid residues at the amino or carboxy termini are considered to be sequences of a single species. Such microheterogeneity is a common feature of manufactured proteins.

The term "specific binding fragment" as used herein with reference to a full-length wild-type mammalian CD80 polypeptide or an IgV or IgC domain thereof means a polypeptide having a subsequence of the IgV and/or IgC domain and specifically binds to mammalian CD28, mammalian PD-L1, and/or mammalian CTLA-4 (e.g., human or murine CD28, PD-L1, and/or CTLA-4) in vitro and/or in vivo. In some embodiments, a specific binding fragment of CD80 IgV or CD80 IgC is at least 60%, 70%, 75%, 80%, 85%, 90%, 95%, 96%, 97%, 98%, or 99% of the sequence length of the full-length wild-type sequence. The sequence of the specific binding fragment can be altered to form a variant CD80.

The term "specifically bind" as used herein refers to the ability of a protein to bind to a target protein under specific binding conditions such that its affinity or avidity is at least 5 times greater than the average affinity or avidity of the same protein for a collection of random peptides or polypeptides of sufficient statistical size, but optionally at least 10, 20, 30, 40, 50, 100, 250 or 500 times greater, or even at least 1000 times greater. A specifically binding protein need not only bind to a single target molecule, but may specifically bind to a non-target molecule due to the similarity of configuration between the target and the non-target (e.g., paralog or ortholog). Those skilled in the art will recognize that specific binding to molecules with the same function in different animal species (i.e., orthologs) or to non-target molecules with substantially similar epitopes to the target molecule (e.g., paralogs) is possible and does not impair the specificity of the binding determined for a statistically valid collection of unique non-targets (e.g., random polypeptides). Thus, due to cross-reactivity, the polypeptide of the present invention can specifically bind to multiple distinct target molecular species.The specific binding between two proteins can be determined using solid-phase ELISA immunoassay or Biacore measurement.Generally, the interaction between two binding proteins has a dissociation constant (Kd) of less than 1×10 ⁻⁵ M, often as low as 1×10 ⁻¹² M.In certain embodiments of the present disclosure, the interaction between two binding proteins has a dissociation constant of 1×10 ⁻⁶ M, 1×10 ⁻⁷ M, 1×10 ⁻⁸ M, 1×10 ⁻⁹ M, 1×10 ^{−10 M or 1×10 −11 M.}

With respect to a mammalian cell expressing a polypeptide, the term "surface expressed" or "surface expression" means that the polypeptide is expressed as a membrane protein. In some embodiments, the membrane protein is a transmembrane protein.

As used herein, "synthetic" refers to a nucleic acid molecule or a polypeptide molecule that is produced by recombinant and/or chemical synthesis methods, e.g., with respect to a synthetic nucleic acid molecule or a synthetic gene or a synthetic peptide.

The term "targeting moiety" as used herein refers to a composition that is covalently or non-covalently bound to or physically encapsulates a polypeptide comprising a variant CD80. The targeting moiety has a specific binding affinity to a desired counter structure, such as a cell surface receptor (e.g., B7 family member PD-L1) or a tumor antigen (e.g., tumor-specific antigen (TSA) or tumor-associated antigen (TAA), e.g., B7-H6). Typically, the desired counter structure is localized on a particular tissue or cell type. Targeting moieties include antibodies, antigen-binding fragments (Fab), variable fragments (Fv) containing _VH and _VL , single chain variable fragments (scFv) containing _VH and _VL linked together in one chain, and other antibody V region fragments, such as Fab', F(ab) ₂ , F(ab') ₂ , dsFv diabodies, nanobodies, soluble receptors, receptor ligands, affinity matured receptors or ligands, and small molecule (<500 daltons) compositions (e.g., specific binding receptor compositions). Targeting moieties can also be covalently or non-covalently bound (attached) to the lipid membrane of liposomes that encapsulate the polypeptides of the invention.

As used herein, the term "transmembrane protein" refers to a membrane protein that substantially or completely spans a lipid bilayer, which is found, for example, in biological membranes, such as mammalian cells, or in artificial constructs, such as liposomes. A transmembrane protein includes a transmembrane domain ("transmembrane domain") that is integrated into the lipid bilayer and whose integration is thermodynamically stable under physiological conditions. The transmembrane domain is generally predictable from the amino acid sequence of the transmembrane domain through a number of commercially available bioinformatics software applications, based on the high hydrophobicity of the transmembrane domain compared to the regions of the protein that interact with the aqueous environment (e.g., cytosol, extracellular fluid). The transmembrane domain is often a hydrophobic alpha helix that spans the membrane. A transmembrane protein may span both layers of the lipid bilayer one or more times. The provided transmembrane immunomodulatory proteins described herein are included in transmembrane proteins. In addition to the transmembrane domain, the transmembrane immunomodulatory proteins of the present invention further include an ectodomain, and in some embodiments, an endodomain.

The term "treatment" or "therapy" of a disease or disorder, as used herein, means slowing, halting, or reversing the progression of the disease or disorder as evidenced by the reduction, arrest, or elimination of any clinical or diagnostic symptoms by administering a therapeutic composition of the invention (e.g., containing an immunomodulatory protein or modified cell), either alone or in combination with another compound as described herein. "Treating" or "treatment" also means reducing the severity of symptoms in an acute or chronic disease or disorder, or reducing the rate of relapse (e.g., as in relapsing or remitting autoimmune disease processes), or reducing inflammation in the case of the inflammatory aspects of an autoimmune disease. As used herein in the context of cancer, the terms "treating" cancer or "inhibiting" cancer or "inhibiting" cancer refer to at least one of the following: a statistically significant reduction in tumor growth rate, cessation of tumor growth, or a reduction in tumor size, mass, metabolic activity, or volume, or a statistically significant improvement in progression-free survival (PFS) or overall survival (OS), as measured by standard criteria, such as, but not limited to, Response Evaluation Criteria for Solid Tumors (RECIST). "Preventing" a disease or disorder or "prevention" of a disease or disorder, as used in the context of the present invention, refers to the administration of an immune-modulating polypeptide or modified cell of the present invention, either alone or in combination with another compound, to prevent the appearance or development of a disease or disorder or some or all of the symptoms of a disease or disorder, or to reduce the likelihood of developing a disease or disorder.

The term "tumor specific antigen" or "TSA" as used herein refers to a counter structure that is present primarily on tumor cells of a mammalian subject, but is not generally found on normal cells of the mammalian subject. A tumor specific antigen need not be present exclusively on tumor cells, but the percentage of cells in a particular mammal that have the tumor specific antigen or the level of the tumor specific antigen on the surface of the tumor is high enough that it can be targeted with an anti-tumor therapeutic (e.g., an immunomodulatory polypeptide of the invention) and provide for the prevention or treatment of the mammal from the effects of the tumor. In some embodiments, in a random statistical sample of cells from a mammal with a tumor, at least 50% of the cells that present the TSA are cancerous. In other embodiments, at least 60%, 70%, 80%, 85%, 90%, 95%, or 99% of the cells that present the TSA are cancerous.

The term "variant" (also "modified" or "mutant"), when used with respect to variant CD80, refers to a CD80 created by human intervention, e.g., mammalian (e.g., human or murine) CD80. A variant CD80 is a polypeptide having an altered amino acid sequence compared to unmodified or wild-type CD80. A variant CD80 is a polypeptide that differs from the wild-type CD80 isoform sequence by one or more amino acid substitutions, deletions, additions, or combinations thereof. For purposes herein, a variant CD80 contains at least one affinity-modified domain whereby one or more of the amino acid differences occur in an IgSF domain (e.g., an IgV domain). A variant CD80 can contain 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30 or more amino acid differences, e.g., amino acid substitutions. A variant CD80 polypeptide generally exhibits at least 50%, 60%, 70%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more sequence identity to a corresponding wild-type or unmodified CD80 (e.g., the sequence of SEQ ID NO:1), its mature sequence, or a portion thereof containing the extracellular domain or the IgSF domain thereof. In some embodiments, the variant CD80 polypeptide exhibits at least 50%, 60%, 70%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or higher sequence identity to the corresponding wild-type or unmodified CD80, including the sequence shown in SEQ ID NO:28, SEQ ID NO:152, or SEQ ID NO:372. Both non-naturally occurring and naturally occurring amino acids are included within the scope of permissible substitutions or additions. The variant CD80 is not limited to any particular method of production, including, for example, de novo chemical synthesis, de novo recombinant DNA technology, or a combination thereof. The variant CD80 of the present invention specifically binds to at least one or more of CD28, PD-L1, and/or CTLA-4 of a mammalian species. In some embodiments, the amino acid sequence change results in a change (i.e., an improvement or decrease) in the binding affinity or avidity to CD28, PD-L1, and/or CTLA-4 compared to unmodified or wild-type CD80 protein. The improvement or decrease in binding affinity or avidity can be determined using well-known binding assays such as flow cytometry. Larsen et al., American Journal of Transplantation, Vol 5: 443-453 (2005). See also Linsley et al., Immunity, 1: 7930801 (1994). The improvement in binding affinity or avidity of variant CD80 for CD28, PD-L1, and/or CTLA-4 can be an improvement to a value at least 5% greater than that of the unmodified or wild-type CD80, and in some embodiments can be an improvement to a value at least 10%, 15%, 20%, 30%, 40%, 50%, 100% greater than that of the unmodified or wild-type CD80 control value. The decrease in binding affinity or avidity of CD80 for CD28, PD-L1, and/or CTLA-4 can be a decrease to a value no greater than 95% of the unmodified or wild-type CD80 control value, and in some embodiments a decrease to no greater than 80%, 70%, 60%, 50%, 40%, 30%, 20%, 10%, 5% or less of the binding affinity or avidity of the unmodified or wild-type CD80 control value, or to an undetectable value. A variant CD80 polypeptide has an altered primary amino acid sequence due to substitution, addition, or deletion of amino acid residues. The term "variant" in the context of variant CD80 polypeptides should not be construed as mandating any particular starting composition or any condition of the method by which the variant CD80 is made. A variant CD80 can be generated, for example, starting from wild-type mammalian CD80 sequence information, then modeled in silico for binding to CD28, PD-L1, and/or CTLA-4, and finally recombinantly or chemically synthesized to generate the variant CD80. As just another example, a variant CD80 can be generated by site-directed mutagenesis of unmodified or wild-type CD80. Thus, a variant CD80 represents a composition, but not necessarily a product produced by any given process. A wide variety of techniques may be employed, including recombinant methods, chemical synthesis, or a combination thereof.

The terms "wild-type" or "natural" or "native" as used herein in connection with biological materials, such as nucleic acid molecules, proteins (e.g., CD80), IgSF members, host cells, etc., refer to those found in nature and unmodified by human intervention.

II. Variant CD80 Polypeptides
Provided herein are variant CD80 polypeptides that have altered (improved or decreased) binding activity or affinity to one or more CD80 cognate binding partners. In some embodiments, the CD80 cognate binding partner is CD28, PD-L1, or CTLA-4. In some embodiments, the variant CD80 polypeptide contains one or more amino acid modifications, such as one or more substitutions (or "mutations" or "exchanges"), deletions, or additions, in the immunoglobulin superfamily (IgSF) domain (IgD) compared to a wild-type or unmodified CD80 polypeptide, or an IgD-containing portion of a wild-type or unmodified CD80, or a specific-binding fragment thereof.

Thus, the variant CD80 polypeptides provided are or include variant IgDs (hereinafter referred to as "vIgD") in which one or more amino acid modifications (e.g., substitutions) are present in IgD.

In some embodiments, the IgD comprises an IgV domain or an IgC (e.g., IgC2) domain, or a specific binding fragment of an IgV domain or an IgC (e.g., IgC2) domain, or a combination thereof. In some embodiments, the IgD can be just an IgV, or a combination of IgV and IgC including the entire extracellular domain (ECD), or any combination of the Ig domains of CD80. Table 2 provides exemplary residues corresponding to the IgV or IgC regions of CD80. In some embodiments, the variant CD80 polypeptide contains an IgV domain or an IgC domain, or a specific binding fragment thereof, with at least one amino acid modification (e.g., substitution) in the IgV domain or the IgC domain, or a specific binding fragment thereof. In some embodiments, the variant CD80 polypeptide contains an IgV domain or a specific binding fragment thereof, with at least one amino acid modification (e.g., substitution) in the IgV domain or the IgC domain, or a specific binding fragment thereof. In some embodiments, the altered IgV or IgC domain is an affinity-modified IgSF domain with altered binding activity or affinity.

In some embodiments, the variant has one more IgSF domain modified compared to the sequence of the unmodified CD80 sequence. In some embodiments, the unmodified CD80 sequence is wild-type CD80. In some embodiments, the unmodified or wild-type CD80 has the sequence of a native CD80 or an orthologue thereof. In some embodiments, the unmodified CD80 is or comprises an extracellular domain (ECD) or a portion thereof of CD80 that contains one or more IgSF domains (see Table 2). In some embodiments, the extracellular domain of the unmodified or wild-type CD80 polypeptide comprises an IgV domain and an IgC domain. However, the variant CD80 polypeptide need not comprise both an IgV domain and an IgC domain. In some embodiments, the variant CD80 polypeptide comprises or consists essentially of an IgV domain or a specific binding fragment thereof. In some embodiments, the variant CD80 polypeptide comprises or consists essentially of an IgC domain or a specific binding fragment thereof. In some embodiments, the variant CD80 is soluble and lacks a transmembrane domain. In some embodiments, the variant CD80 further comprises a transmembrane domain, and in some cases, a cytoplasmic domain.

In some embodiments, the wild-type or unmodified CD80 polypeptide is a mammalian CD80 polypeptide, such as, but not limited to, a human, mouse, cynomolgus monkey, or rat CD80 polypeptide. In some embodiments, the wild-type or unmodified CD80 sequence is human.

In some embodiments, the wild-type or unmodified CD80 polypeptide (i) has the amino acid sequence set forth in SEQ ID NO:1 or a mature form thereof lacking a signal sequence, (ii) has an amino acid sequence exhibiting at least about 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% sequence identity to SEQ ID NO:1 or a mature form thereof, or (iii) is a portion of (i) or (ii) that contains an IgV domain or an IgC domain or a specific binding fragment thereof.

In some embodiments, the wild-type or unmodified CD80 polypeptide is or comprises the extracellular domain of CD80 or a portion thereof. For example, in some embodiments, the unmodified or wild-type CD80 polypeptide comprises the amino acid sequence set forth in SEQ ID NO:28, or an orthologue thereof. For example, the unmodified or wild-type CD80 polypeptide can (i) comprise the amino acid sequence set forth in SEQ ID NO:28, (ii) comprise an amino acid sequence having at least about 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% sequence identity to SEQ ID NO:28, or (iii) be a specific binding fragment of (i) or (ii) comprising the IgV domain or the IgC domain. In some embodiments, the wild-type or unmodified extracellular domain of CD80 can bind to one or more CD80 cognate binding proteins, such as one or more of CD28, PD-L1, or CTLA-4.

In some embodiments, the wild-type or unmodified CD80 polypeptide contains an IgV domain or an IgC domain or a specific binding fragment thereof. In some embodiments, the IgV domain of the wild-type or unmodified CD80 polypeptide comprises the amino acid sequence set forth in SEQ ID NO: 152 or 372, or an orthologue thereof. For example, the IgV domain of the unmodified or wild-type CD80 polypeptide can (i) contain the amino acid sequence set forth in SEQ ID NO: 152 or 372, (ii) contain an amino acid sequence having at least about 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% sequence identity to SEQ ID NO: 152 or 372, or (iii) is a specific binding fragment of (i) or (ii). In some embodiments, the wild-type or unmodified IgV domain can bind to one or more CD80 cognate binding proteins, such as one or more of CD28, PD-L1, or CTLA-4.

In some embodiments, the IgC domain of a wild-type or unmodified CD80 polypeptide comprises the amino acid sequence set forth as residues 145-230 or 154-232 of SEQ ID NO:1, or an orthologue thereof. For example, the IgC domain of an unmodified or wild-type CD80 polypeptide can (i) contain the amino acid sequence set forth as residues 145-230 or 154-232 of SEQ ID NO:1, (ii) contain an amino acid sequence having at least about 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% sequence identity to residues 145-230 or 154-232 of SEQ ID NO:1, or (iii) is a specific binding fragment of (i) or (ii). In some embodiments, the wild-type or unmodified IgC domain is capable of binding to one or more CD80 cognate binding proteins.

In some embodiments, the wild-type or unmodified CD80 polypeptide contains a specific binding fragment of CD80, e.g., a specific binding fragment of an IgV domain or an IgC domain. In some embodiments, the specific binding fragment can bind to CD28, PD-L1, and/or CTLA 4. The specific binding fragment can have a length of at least 50 amino acids, e.g., at least 60, 70, 80, 90, 100, or 110 amino acids. In some embodiments, the specific binding fragment of an IgV domain contains an amino acid sequence that is at least about 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% of the length of the IgV domain set forth as amino acids 35-135 or 37-138 of SEQ ID NO:1. In some embodiments, the specific binding fragment of the IgC domain comprises an amino acid sequence that is at least about 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% of the length of the IgC domain set forth as amino acids 145-230 or 154-232 of SEQ ID NO:1.

In some embodiments, the variant CD80 polypeptide comprises an ECD domain or a portion thereof comprising one or more affinity modified IgSF domains. In some embodiments, the variant CD80 polypeptide comprises an IgV domain or an IgC domain, or a specific binding fragment of an IgV domain or a specific binding fragment of an IgC domain, where at least one of the IgV or IgC domains contains one or more amino acid modifications (e.g., substitutions). In some embodiments, the variant CD80 polypeptide comprises an IgV domain and an IgC domain, or a specific binding fragment of an IgV domain and a specific binding fragment of an IgC domain. In some embodiments, the variant CD80 polypeptide comprises a full-length IgV domain. In some embodiments, the variant CD80 polypeptide comprises a full-length IgC domain. In some embodiments, the variant CD80 polypeptide comprises a specific binding fragment of an IgV domain. In some embodiments, the variant CD80 polypeptide comprises a specific binding fragment of an IgC domain. In some embodiments, the variant CD80 polypeptide comprises a full-length IgV domain and a full-length IgC domain. In some embodiments, the variant CD80 polypeptide comprises a full-length IgV domain and a specific binding fragment of the IgC domain. In some embodiments, the variant CD80 polypeptide comprises a specific binding fragment of the IgV domain and a full-length IgC domain. In some embodiments, the variant CD80 polypeptide comprises a specific binding fragment of the IgV domain and a specific binding fragment of the IgC domain.

In any of such embodiments, the one or more amino acid modifications (e.g., substitutions) of the variant CD80 polypeptide can be located in any one or more of the CD80 polypeptide domains. For example, in some embodiments, the one or more amino acid modifications (e.g., substitutions) are located in the extracellular domain of the variant CD80 polypeptide. In some embodiments, the one or more amino acid modifications (e.g., substitutions) are located in the IgV domain or a specific binding fragment of the IgV domain. In some embodiments, the one or more amino acid modifications (e.g., substitutions) are located in the IgC domain or a specific binding fragment of the IgC domain.

Generally, each of the various attributes of the polypeptides are disclosed separately below (e.g., soluble and membrane-bound polypeptides, affinity of CD80 for CD28, PD-L1, and CTLA-4, number of differences per polypeptide chain, number of linked polypeptide chains, number and nature of amino acid changes per variant CD80, etc.). However, as will be apparent to one of skill in the art, any particular polypeptide may include a combination of these independent attributes. It will be understood that references to amino acids, including references to specific sequences shown as SEQ ID NOs used to describe the domain organization of the IgSF domains, are for illustrative purposes and are not meant to limit the scope of the embodiments provided. It will be understood that the description of the polypeptides and their domains is theoretically derived based on homology analysis and alignment with similar molecules. Thus, the exact locus may vary and is not necessarily the same from protein to protein. Thus, a particular IgSF domain, e.g., a particular IgV domain or IgC domain, may be several amino acids longer or shorter (e.g., 1, 2, 3, or 4).

Furthermore, the various aspects of the invention as discussed below are often provided within the meaning of defined terms as disclosed above. Therefore, the aspects described in a particular definition should be construed as being incorporated by reference when the defined terms are utilized in the discussion of the various aspects and attributes described herein. Thus, the headings, the order of presentation of the various aspects and embodiments, and the separate disclosure of each independent attribute are not meant to limit the scope of the disclosure.

Exemplary Modifications Provided herein are variant CD80 polypeptides that contain at least one affinity-modified IgSF domain (e.g., IgV or IgC) or specific binding fragment thereof, as compared to the IgSF domain contained in a wild-type or unmodified CD80 polypeptide, and that have an altered (improved or decreased) binding activity or affinity for one or more ligands, CD28, PD-L1, or CTLA-4, as compared to the wild-type or unmodified CD80 polypeptide. In some embodiments, the variant CD80 polypeptide has a binding affinity for CD28, PD-L1, and/or CTLA-4 that is different from that of a wild-type or unmodified CD80 polypeptide control sequence, as determined, for example, by solid-phase ELISA immunoassay, flow cytometry, or Biacore assay. In some embodiments, the variant CD80 polypeptide has an improved binding affinity for CD28, PD-L1, and/or CTLA-4. In some embodiments, the variant CD80 polypeptide has a reduced binding affinity for CD28, PD L1, and/or CTLA-4 compared to a wild-type or unmodified CD80 polypeptide. CD28, PD-L1, and/or CTLA-4 can be mammalian proteins, e.g., human proteins or murine proteins.

The binding affinity for each of the cognate binding partners is independent; i.e., in some embodiments, the variant CD80 polypeptide has improved binding affinity for one, two or three of CD28, PD-L1, and CTLA-4, and/or decreased binding affinity for one, two or three of CD28, PD-L1, and CTLA-4, compared to a wild-type or unmodified CD80 polypeptide.

In some embodiments, the variant CD80 polypeptide has improved binding affinity for CD28 compared to wild-type or unmodified CD80 polypeptides. In some embodiments, the variant CD80 polypeptide has improved binding affinity for PD-L1 compared to wild-type or unmodified CD80 polypeptides. In some embodiments, the variant CD80 polypeptide has improved binding affinity for CTLA-4 compared to wild-type or unmodified CD80 polypeptides. In some embodiments, the variant CD80 polypeptide has reduced binding affinity for CD28 compared to wild-type or unmodified CD80 polypeptides. In some embodiments, the variant CD80 polypeptide has reduced binding affinity for PD-L1 compared to wild-type or unmodified CD80 polypeptides. In some embodiments, the variant CD80 polypeptide has reduced binding affinity for CTLA-4 compared to wild-type or unmodified CD80 polypeptides.

In some embodiments, the variant CD80 polypeptide has improved binding affinity for CD28 and PD-L1 compared to wild-type or unmodified CD80 polypeptides. In some embodiments, the variant CD80 polypeptide has improved binding affinity for CD28 and reduced binding affinity for PD-L1 compared to wild-type or unmodified CD80 polypeptides. In some embodiments, the variant CD80 polypeptide has reduced binding affinity for CD28 and PD-L1 compared to wild-type or unmodified CD80 polypeptides. In some embodiments, the variant CD80 polypeptide has reduced binding affinity for CD28 and improved binding affinity for PD-L1 compared to wild-type or unmodified CD80 polypeptides.

In some embodiments, the variant CD80 polypeptide has improved binding affinity for CD28 and CTLA-4 compared to wild-type or unmodified CD80 polypeptides. In some embodiments, the variant CD80 polypeptide has improved binding affinity for CD28 and reduced binding affinity for CTLA-4 compared to wild-type or unmodified CD80 polypeptides. In some embodiments, the variant CD80 polypeptide has reduced binding affinity for CD28 and CTLA-4 compared to wild-type or unmodified CD80 polypeptides. In some embodiments, the variant CD80 polypeptide has reduced binding affinity for CD28 and improved binding affinity for CTLA-4 compared to wild-type or unmodified CD80 polypeptides.

In some embodiments, the variant CD80 polypeptide has improved binding affinity for PD-L1 and CTLA-4 compared to wild-type or unmodified CD80 polypeptides. In some embodiments, the variant CD80 polypeptide has improved binding affinity for PD-L1 and reduced binding affinity for CTLA-4 compared to wild-type or unmodified CD80 polypeptides. In some embodiments, the variant CD80 polypeptide has reduced binding affinity for PD-L1 and CTLA-4 compared to wild-type or unmodified CD80 polypeptides. In some embodiments, the variant CD80 polypeptide has reduced binding affinity for PD-L1 and improved binding affinity for CTLA-4 compared to wild-type or unmodified CD80 polypeptides.

In some embodiments, the variant CD80 polypeptide has improved binding affinity for CD28, PD-L1, and CTLA-4 compared to wild-type or unmodified CD80 polypeptides. In some embodiments, the variant CD80 polypeptide has improved binding affinity for CD28 and PD-L1, and reduced binding affinity for CTLA-4 compared to wild-type or unmodified CD80 polypeptides. In some embodiments, the variant CD80 polypeptide has improved binding affinity for CD28 and CTLA-4, and reduced binding affinity for PD-L1 compared to wild-type or unmodified CD80 polypeptides. In some embodiments, the variant CD80 polypeptide has reduced binding affinity for CD28 and PD-L1, and improved binding affinity for CTLA-4 compared to wild-type or unmodified CD80 polypeptides. In some embodiments, the variant CD80 polypeptide has reduced binding affinity for CD28 and PD-L1, and improved binding affinity for CTLA-4 compared to wild-type or unmodified CD80 polypeptides. In some embodiments, the variant CD80 polypeptide has improved binding affinity for CD28 and reduced binding affinity for PD-L1 and CTLA-4 compared to wild-type or unmodified CD80 polypeptides. In some embodiments, the variant CD80 polypeptide has reduced binding affinity for CD28, CTLA-4, and PD-L1 compared to wild-type or unmodified CD80 polypeptides. In some embodiments, the variant CD80 polypeptide has reduced binding affinity for CD28 and improved binding affinity for PD-L1 and CTLA-4 compared to wild-type or unmodified CD80 polypeptides.

In some embodiments, a variant CD80 polypeptide having improved or higher binding affinity to CD28, PD-L1, and/or CTLA-4 will have at least about 5%, e.g., at least about 10%, 15%, 20%, 25%, 35%, or 50% improved binding affinity to CD28, PD-L1, and/or CTLA-4 compared to a wild-type or unmodified CD80 polypeptide control. In some embodiments, the improved binding affinity is 1.2-fold, 1.5-fold, 2-fold, 3-fold, 4-fold, 5-fold, 6-fold, 7-fold, 8-fold, 9-fold, 10-fold, 20-fold, 30-fold, 40-fold, or 50-fold greater than a wild-type or unmodified CD80 polypeptide. In such examples, the wild-type or unmodified CD80 polypeptide has the same sequence as the variant CD80 polypeptide, except that it does not contain one or more amino acid modifications (e.g., substitutions).

In some embodiments, a variant CD80 polypeptide having reduced or decreased binding affinity for CD28, PD-L1, and/or CTLA-4 will have at least a 5% decrease in binding affinity, e.g., at least about a 10%, 15%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, or more, for CD28, PD-L1, and/or CTLA-4, as compared to a wild-type or unmodified CD80 polypeptide control. In some embodiments, the decrease in binding affinity is greater than a 1.2-fold, 1.5-fold, 2-fold, 3-fold, 4-fold, 5-fold, 6-fold, 7-fold, 8-fold, 9-fold, 10-fold, 20-fold, 30-fold, 40-fold, or 50-fold decrease as compared to a wild-type or unmodified CD80 polypeptide. In such examples, the wild-type or unmodified CD80 polypeptide has the same sequence as the variant CD80 polypeptide, except that it does not contain one or more amino acid modifications (e.g., substitutions).

In some embodiments, the equilibrium dissociation constant (K _d ) of any of the aforementioned embodiments for CD28, PD-L1, and/or CTLA-4 may be less than 1×10 ⁻⁵ M, 1×10 ⁻⁶ M, 1×10 ⁻⁷ M, 1×10 ⁻⁸ M, 1×10 ⁻⁹ M, 1×10 ⁻¹⁰ M or 1×10 ⁻¹¹ M, or 1×10 ⁻¹² M.

A wild-type or unmodified CD80 sequence does not necessarily have to be used as the starting composition to generate the variant CD80 polypeptides described herein. Thus, the use of the term "substitution" does not imply that the provided embodiments are limited to a particular method of making the variant CD80 polypeptide. Variant CD80 polypeptides can be produced, for example, by de novo peptide synthesis and therefore do not necessarily require a "substitution" in the sense of altering a codon to code for the substitution. This principle also extends to the terms "addition" and "deletion" of amino acid residues, which also do not imply a particular method of production. The means by which variant CD80 polypeptides are designed or made is not limited to any particular method. However, in some embodiments, nucleic acids encoding wild-type or unmodified CD80 are mutagenized from wild-type or unmodified CD80 genetic material and screened for the desired specific binding affinity and/or induction of IFN-γ expression or other functional activity according to the methods disclosed in the Examples or other methods known to those of skill in the art. In some embodiments, variant CD80 polypeptides are synthesized de novo using protein or nucleic acid sequences available in a number of publicly available databases, which are then subsequently screened. The National Center for Biotechnology Information provides such information, and its website is publicly accessible via the internet, as is the UniProtKB database, as discussed above.

Unless otherwise specified, as used throughout this disclosure, amino acid substitutions are designated by amino acid position numbers that correspond to the numbering of the positions in the unmodified ECD sequence shown in SEQ ID NO:28, or the positions in the unmodified IgV sequence shown in SEQ ID NO:152 or 372 (containing residues 1-101 or 1-107, respectively, of SEQ ID NO:28, depending on annotation and convention), as applicable, as follows:

It is within the level of ordinary skill in the art to identify the corresponding position of a modification (e.g., amino acid substitution) in a CD80 polypeptide (including a portion thereof containing its IgSF domain (e.g., IgV)) by alignment to a reference sequence having, for example, SEQ ID NO:28 or SEQ ID NO:152 or SEQ ID NO:372. In descriptions of modifications throughout this disclosure, the amino acid position is centered, the corresponding unmodified (e.g., wild-type) amino acid is listed before the number, and the amino acid substitution of the identified variant is listed after the number. If the modification is a deletion at that position, it is indicated as "del" and if the modification is an insertion at that position, it is indicated as "ins."

In some embodiments, the variant CD80 polypeptide has one or more amino acid modifications (e.g., substitutions) in a wild-type or unmodified CD80 sequence. The one or more amino acid modifications (e.g., substitutions) can be in an ectodomain (e.g., extracellular domain) of a wild-type or unmodified CD80 sequence. In some embodiments, the one or more amino acid modifications (e.g., substitutions) are in an IgV domain or a specific binding fragment thereof. In some embodiments, the one or more amino acid modifications (e.g., substitutions) are in an IgC domain or a specific binding fragment thereof. In some embodiments of the variant CD80 polypeptide, some of the one or more amino acid modifications (e.g., substitutions) are in an IgV domain or a specific binding fragment thereof and some of the one or more amino acid modifications (e.g., substitutions) are in an IgC domain or a specific binding fragment thereof.

In some embodiments, the variant CD80 polypeptide has up to 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20 amino acid modifications (e.g., substitutions). The modifications (e.g., substitutions) can be in the IgV domain or the IgC domain. In some embodiments, the variant CD80 polypeptide has up to 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20 amino acid modifications (e.g., substitutions) in the IgV domain or a specific binding fragment thereof. In some embodiments, the variant CD80 polypeptide has up to 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20 amino acid modifications (e.g., substitutions) in the IgC domain or specific binding fragment thereof. In some embodiments, the variant CD80 polypeptide has at least about 85%, 86%, 86%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity to a wild-type or unmodified CD80 polypeptide or specific binding fragment thereof, e.g., to an amino acid sequence of SEQ ID NO:28, 152, or 372.

In some embodiments, the variant CD80 polypeptide is at position 4, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 20, 21, 22, 24, 25, 27, 28, 29, 30, 31, 33, 36, 37, 38, 40, 41, 42, 43, 44, 47, 48, 50, 52, 53, 54, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 72, 74, 76, 77, 80, 81, 83, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 99, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146 160, 162, 164, 168, 169, 174, 175, 177, 178, 183, 178, 185, 188, 190, 192, 193, or 199. In some embodiments, such variant CD80 polypeptides have altered binding affinity to one or more of CD28, PD-L1, or CTLA-4 compared to a wild-type or unmodified CD80 polypeptide. For example, in some embodiments, the variant CD80 polypeptides have improved binding affinity to CD28, PD-L1, and/or CTLA-4 compared to a wild-type or unmodified CD80 polypeptide. In some embodiments, the variant CD80 polypeptides have reduced binding affinity to CD28, PD-L1, and/or CTLA-4 compared to a wild-type or unmodified CD80 polypeptide.

から選択される1つまたは複数のアミノ酸置換を有する。いくつかの態様において、バリアントCD80ポリペプチドは、

またはその保存的アミノ酸置換からなる群より選択される1つまたは複数のアミノ酸置換を有する。 In some embodiments, the variant CD80 polypeptide comprises:

In some embodiments, the variant CD80 polypeptide has one or more amino acid substitutions selected from:

or conservative amino acid substitutions thereof.

である。 In some embodiments, the one or more amino acid modifications, e.g., substitutions,

It is.

In some embodiments, the variant CD80 polypeptide is at position 4, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 20, 21, 22, 24, 25, 27, 28, 29, 31, 33, 36, 37, 38, 40, 41, 42, 43, 44, 47, 48, 50, 52, 53, 54, 59, 60, 61, 62, 63, 64, 66, 67, 68, 69, 70, 72, 74, 76, 77, 80, 81, 83, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 99, 102, 104, 106, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 1 , 103, 104, 108, 109, 110, 115, 116, 117, 118, 120, 121, 122, 126, 127, 128, 129, 130, 133, 137, 140, 142, 143, 144, 148, 149, 152, 154, 160, 162, 168, 169, 174, 175, 177, 178, 183, 178, 188, 190, 192, or 193. In some embodiments, such variant CD80 polypeptides have altered binding affinity for one or more of CD28, PD-L1, or CTLA-4 as compared to a wild-type or unmodified CD80 polypeptide. For example, in some embodiments, the variant CD80 polypeptide has improved binding affinity to CD28, PD-L1, and/or CTLA-4 compared to a wild-type or unmodified CD80 polypeptide. In some embodiments, the variant CD80 polypeptide has reduced binding affinity to CD28, PD-L1, and/or CTLA-4 compared to a wild-type or unmodified CD80 polypeptide.

から選択される1つまたは複数のアミノ酸置換を有する。いくつかの態様において、バリアントCD80ポリペプチドは、

またはその保存的アミノ酸置換からなる群より選択される1つまたは複数のアミノ酸置換を有する。 In some embodiments, the variant CD80 polypeptide comprises:

In some embodiments, the variant CD80 polypeptide has one or more amino acid substitutions selected from:

or conservative amino acid substitutions thereof.

In some embodiments, the variant CD80 polypeptide has one or more amino acid modifications (e.g., substitutions) in unmodified CD80 or a specific binding fragment thereof corresponding to positions 30, 65, 107, 114, 164, 185, or 199 based on the numbering of SEQ ID NO:28. In some embodiments, the variant CD80 polypeptide has one or more amino acid substitutions selected from I30V, N48D, L65H, E99D, D107N, S114T, I126V, N144D, Y164H, and T199S. In some embodiments, the variant CD80 polypeptide has one or more amino acid substitutions selected from the group consisting of N48D, L65H, E99D, D107N, S114T, I126V, N144D, Y164H, K185E, T199S, or conservative amino acid substitutions thereof.

A conservative amino acid substitution is any amino acid that belongs to the same class of amino acids as the substituted amino acid, other than the wild-type or unmodified amino acid. The classes of amino acids are aliphatic (glycine, alanine, valine, leucine, and isoleucine), hydroxyl or sulfur-containing (serine, cysteine, threonine, and methionine), cyclic (proline), aromatic (phenylalanine, tyrosine, tryptophan), basic (histidine, lysine, and arginine), and acidic/amide (aspartic acid, glutamic acid, asparagine, and glutamine). Thus, for example, conservative amino acid substitutions for a V4M substitution include V4S, V4C, and V4T amino acid substitutions.

In some embodiments, the variant CD80 polypeptide comprises any of the substitutions (mutations) listed in Table 1. Table 1 also provides exemplary sequences by reference to SEQ ID NOs for the extracellular domain (ECD) or IgV domain of wild-type CD80 or exemplary variant CD80 polypeptides. As indicated, the exact locus or residues corresponding to a given domain may vary depending, for example, on the method used to identify or classify the domain. In some cases, adjacent amino acids N-terminal and/or C-terminal to a given domain (e.g., IgV) may also be included in the sequence of the variant IgSF polypeptide, for example, to ensure proper folding of the domain when expressed. Thus, it is understood that the exemplary SEQ ID NOs in Table 1 should not be construed as limiting. For example, a particular domain (e.g., IgV domain) of a variant CD80 polypeptide may be several amino acids longer or shorter than the amino acid sequence shown in the respective SEQ ID NO, e.g., 1 to 10 amino acids, e.g., 1, 2, 3, 4, 5, 6, or 7 amino acids longer or shorter.

In some embodiments, the variant CD80 polypeptide comprises any of the extracellular domain (ECD) sequences listed in Table 1 (i.e., any one of SEQ ID NOs:55-108, 280-346, 414-475). In some embodiments, the variant CD80 polypeptide comprises a polypeptide sequence that exhibits at least 90% identity, at least 91% identity, at least 92% identity, at least 93% identity, at least 94% identity, at least 95% identity, e.g., at least 96% identity, 97% identity, 98% identity, or 99% identity to any of the extracellular domain (ECD) sequences listed in Table 1 (i.e., any one of SEQ ID NOs:55-108, 280-346, 414-475), and contains an amino acid modification (e.g., a substitution) that is not present in wild-type or unmodified CD80. In some embodiments, the variant CD80 polypeptide comprises a specific binding fragment of any of the extracellular domain (ECD) sequences listed in Table 1 (i.e., any one of SEQ ID NOs:55-108, 280-346, 414-475) and contains an amino acid modification (e.g., a substitution) that is not present in wild-type or unmodified CD80. In some embodiments, the variant CD80 polypeptide comprises any of the IgV sequences listed in Table 1 (i.e., any one of SEQ ID NOs:153-195, 347, 373-386, 476-477). In some embodiments, the variant CD80 polypeptide comprises a polypeptide sequence that exhibits at least 90% identity, at least 91% identity, at least 92% identity, at least 93% identity, at least 94% identity, at least 95% identity, such as at least 96% identity, 97% identity, 98% identity, or 99% identity to any of the IgV sequences listed in Table 1 (i.e., any one of SEQ ID NOs:153-195, 347, 373-386, 476-477), and contains an amino acid modification (e.g., a substitution) that is not present in wild-type or unmodified CD80. In some embodiments, the variant CD80 polypeptide comprises a specific binding fragment of any of the IgV sequences listed in Table 1 (i.e., any one of SEQ ID NOs:153-195, 347, 373-386, 476-477) and contains an amino acid substitution (e.g., a substitution) that is not present in wild-type or unmodified CD80.

Table 1 also provides exemplary sequences by reference to SEQ ID NOs for the extracellular domain (ECD) or IgV domain of wild-type CD80 or exemplary variant CD80 polypeptides. As indicated, the exact locus or residues corresponding to a given domain may vary depending, for example, on the method used to identify or classify the domain. Also, in some cases, adjacent N-terminal and/or C-terminal amino acids of a given domain (e.g., ECD) may also be included in the sequence of the variant IgSF polypeptide, for example, to ensure proper folding of the domain when expressed. Thus, it is understood that the exemplary SEQ ID NOs in Table 1 should not be construed as limiting. For example, a particular domain (e.g., IgV domain) of a variant CD80 polypeptide may be several amino acids longer or shorter than the amino acid sequence shown in the respective SEQ ID NO, e.g., 1 to 10 amino acids, e.g., 1, 2, 3, 4, 5, 6, or 7 amino acids longer or shorter.

Table 1. Exemplary variant CD80 polypeptides

から選択される1つまたは複数のアミノ酸置換を有する。いくつかの態様において、バリアントCD80ポリペプチドは、

またはその保存的アミノ酸置換から選択される1つまたは複数のアミノ酸置換を有する。いくつかの態様において、1つまたは複数のアミノ酸置換は、

である。いくつかの態様において、バリアントCD80ポリペプチドは、CD28のエクトドメインに対する親和性および/またはCD28に対する選択性が、例えばSEQ ID NO:28、152、または372に示される配列を含む野生型または未改変CD80ポリペプチドと比較して向上している。 In some embodiments, the variant CD80 polypeptide has one or more amino acid modifications (e.g., substitutions) corresponding to positions 12, 18, 20, 29, 31, 36, 40, 41, 43, 52, 59, 60, 63, 67, 70, 77, 81, 87, 88, 89, 90, 91, 92, 93, 107, 109, 114, 117, 118, 120, 122, 127, 130, 144, 169, 178, or 199 of SEQ ID NO:28.

In some embodiments, the variant CD80 polypeptide has one or more amino acid substitutions selected from:

or conservative amino acid substitutions thereof. In some embodiments, the one or more amino acid substitutions are

In some embodiments, the variant CD80 polypeptide has improved affinity for and/or selectivity for CD28 ectodomain, as compared to a wild-type or unmodified CD80 polypeptide comprising, e.g., a sequence as set forth in SEQ ID NO:28, 152, or 372.

からなる群より選択される1つまたは複数のアミノ酸置換を有する。いくつかの態様において、バリアントCD80ポリペプチドは、

ならびにその保存的アミノ酸置換からなる群より選択される1つまたは複数のアミノ酸置換を有する。 In some embodiments, the variant CD80 polypeptide has one or more amino acid modifications (e.g., substitutions) corresponding to positions 12, 18, 21, 22, 28, 29, 31, 33, 36, 38, 40, 41, 42, 43, 47, 48, 59, 64, 67, 68, 70, 77, 81, 85, 87, 88, 89, 90, 91, 92, 93, 94, 97, 104, 109, 115, 117, 118, 120, 122, 126, 130, 133, 140, 144, 148, 149, 168, 169, 177, 183, 188, or 193 of SEQ ID NO:28.

In some embodiments, the variant CD80 polypeptide has one or more amino acid substitutions selected from the group consisting of:

and conservative amino acid substitutions thereof.

である。いくつかの態様において、バリアントCD80ポリペプチドは、PD-L1に対する親和性および/またはPD-L1に対する選択性が、例えばSEQ ID NO:28、152、または372に示される配列を含む野生型または未改変CD80ポリペプチドと比較して向上している。 In some embodiments, the one or more amino acid substitutions are:

In some embodiments, the variant CD80 polypeptide has improved affinity for and/or selectivity for PD-L1 compared to a wild-type or unmodified CD80 polypeptide comprising, for example, a sequence as shown in SEQ ID NO:28, 152, or 372.

からなる群より選択される1つまたは複数のアミノ酸置換を有する。いくつかの態様において、バリアントCD80ポリペプチドは、

ならびにその保存的変異からなる群より選択される1つまたは複数のアミノ酸置換を有する。いくつかの態様において、1つまたは複数のアミノ酸置換は、

である。いくつかの態様において、バリアントCD80ポリペプチドは、CD28のエクトドメインおよびPD-L1のエクトドメインに対する親和性が、例えばSEQ ID NO:28、152、または372に示される配列を含む野生型または未改変CD80ポリペプチドと比較して向上している。 In some embodiments, the variant CD80 polypeptide has one or more amino acid modifications (e.g., substitutions) corresponding to positions 12, 18, 36, 40, 43, 59, 77, 88, 89, 90, 91, 92, 93, 109, 118, 122, or 177 of SEQ ID NO:28.

In some embodiments, the variant CD80 polypeptide has one or more amino acid substitutions selected from the group consisting of:

and conservative variations thereof. In some embodiments, the one or more amino acid substitutions are selected from the group consisting of:

In some embodiments, the variant CD80 polypeptide has improved affinity for the ectodomain of CD28 and the ectodomain of PD-L1 compared to a wild-type or unmodified CD80 polypeptide comprising, e.g., a sequence as set forth in SEQ ID NO:28, 152, or 372.

からなる群より選択される1つまたは複数のアミノ酸置換を有する。いくつかの態様において、バリアントCD80ポリペプチドは、

ならびにその保存的アミノ酸置換からなる群より選択される1つまたは複数のアミノ酸置換を有する。いくつかの態様において、1つまたは複数のアミノ酸置換は、

である。いくつかの態様において、バリアントCD80ポリペプチドは、CTLA-4のエクトドメインに対する親和性および/または選択性が、例えばSEQ ID NO:28、152、または372に示される配列を含む野生型または未改変CD80ポリペプチドと比較して向上している。いくつかの態様において、バリアントCD80ポリペプチドは、SEQ ID NO:28の位置36、40、52、70、88、89、90、91、92、93、107、118、120、130、144、または169に対応する、1つまたは複数のアミノ酸改変（例えば、置換）を有する。いくつかの態様において、バリアントCD80ポリペプチドは、

からなる群より選択される1つまたは複数のアミノ酸置換を有する。いくつかの態様において、バリアントCD80ポリペプチドは、

ならびにその保存的アミノ酸置換からなる群より選択される1つまたは複数のアミノ酸置換を有する。いくつかの態様において、1つまたは複数のアミノ酸置換は、

である。いくつかの態様において、バリアントCD80ポリペプチドは、CD28のエクトドメインおよびCTLA-4のエクトドメインに対する親和性が、例えばSEQ ID NO:28、152、または372に示される配列を含む野生型または未改変CD80ポリペプチドと比較して向上している。 In some embodiments, the variant CD80 polypeptide has one or more amino acid modifications (e.g., substitutions) corresponding to positions 4, 29, 31, 36, 40, 41, 52, 67, 68, 70, 87, 88, 89, 90, 91, 92, 93, 107, 109, 110, 118, 120, 130, 144, or 169 of SEQ ID NO:28.

In some embodiments, the variant CD80 polypeptide has one or more amino acid substitutions selected from the group consisting of:

and conservative amino acid substitutions thereof. In some embodiments, the one or more amino acid substitutions are selected from the group consisting of:

In some embodiments, the variant CD80 polypeptide has improved affinity and/or selectivity for the ectodomain of CTLA-4, as compared to a wild-type or unmodified CD80 polypeptide comprising, e.g., a sequence as set forth in SEQ ID NO:28, 152, or 372. In some embodiments, the variant CD80 polypeptide has one or more amino acid modifications (e.g., substitutions) corresponding to positions 36, 40, 52, 70, 88, 89, 90, 91, 92, 93, 107, 118, 120, 130, 144, or 169 of SEQ ID NO:28. ...

In some embodiments, the variant CD80 polypeptide has one or more amino acid substitutions selected from the group consisting of:

and conservative amino acid substitutions thereof. In some embodiments, the one or more amino acid substitutions are

In some embodiments, the variant CD80 polypeptide has improved affinity for the ectodomain of CD28 and the ectodomain of CTLA-4, as compared to a wild-type or unmodified CD80 polypeptide comprising, e.g., a sequence as set forth in SEQ ID NO:28, 152, or 372.

からなる群より選択される1つまたは複数のアミノ酸置換を有する。いくつかの態様において、バリアントCD80ポリペプチドは、

およびその保存的アミノ酸置換からなる群より選択される1つまたは複数のアミノ酸置換を有する。いくつかの態様において、1つまたは複数のアミノ酸置換は、

である。いくつかの態様において、バリアントCD80ポリペプチドは、PD-L1のエクトドメインおよびCTLA-4のエクトドメインに対する親和性が、例えばSEQ ID NO:28、152、または372に示される配列を含む野生型または未改変CD80ポリペプチドと比較して向上している。 In some embodiments, the variant CD80 polypeptide has one or more amino acid modifications (e.g., substitutions) corresponding to positions 29, 31, 36, 40, 41, 67, 70, 87, 88, 89, 90, 91, 92, 93, 109, 118, 120, 122, 130, or 178 of SEQ ID NO:28.

In some embodiments, the variant CD80 polypeptide has one or more amino acid substitutions selected from the group consisting of:

and conservative amino acid substitutions thereof. In some embodiments, the one or more amino acid substitutions are

In some embodiments, the variant CD80 polypeptide has improved affinity for the ectodomain of PD-L1 and the ectodomain of CTLA-4 compared to a wild-type or unmodified CD80 polypeptide comprising, e.g., a sequence as shown in SEQ ID NO:28, 152, or 372.

である。いくつかの態様において、バリアントCD80ポリペプチドは、CD28のエクトドメイン、PD-L1のエクトドメイン、およびCTLA-4のエクトドメインに対する親和性が、例えばSEQ ID NO:28、152、または372に示される配列を含む野生型または未改変CD80ポリペプチドと比較して向上している。 In some embodiments, the variant CD80 polypeptide has one or more amino acid modifications (e.g., substitutions) corresponding to positions 36, 40, 88, 89, 90, 91, 92, or 93 of SEQ ID NO:28. In some embodiments, the variant CD80 polypeptide has one or more amino acid substitutions selected from the group consisting of K36G, L40M, E88D, K89R, D90K, A91G, F92Y, and K93R. In some embodiments, the variant CD80 polypeptide has one or more amino acid substitutions selected from the group consisting of K36G, L40M, E88D, K89R, D90K, A91G, F92Y, K93R, and conservative amino acid substitutions thereof. In some embodiments, the one or more amino acid substitutions are

In some embodiments, the variant CD80 polypeptide has improved affinity for the CD28 ectodomain, the PD-L1 ectodomain, and the CTLA-4 ectodomain, as compared to a wild-type or unmodified CD80 polypeptide comprising, e.g., a sequence as set forth in SEQ ID NO:28, 152, or 372.

からなる群より選択される1つまたは複数のアミノ酸置換を有する。いくつかの態様において、バリアントCD80ポリペプチドは、

ならびにその保存的アミノ酸置換からなる群より選択される1つまたは複数のアミノ酸置換を有する。いくつかの態様において、1つまたは複数のアミノ酸置換は、

である。いくつかの態様において、1つまたは複数のアミノ酸置換は、

である。いくつかの態様において、バリアントCD80ポリペプチドは、例えばSEQ ID NO:28、152、または372に示される配列を含む野生型または未改変CD80ポリペプチドと比較して、CD28のエクトドメインまたはPD-L1のエクトドメインに対する親和性が向上しており、かつCTLA-4のエクトドメインに対する親和性が低下している。 In some embodiments, the variant CD80 polypeptide has one or more amino acid modifications (e.g., substitutions) corresponding to positions 12, 18, 29, 31, 33, 36, 38, 40, 41, 42, 43, 47, 48, 63, 59, 64, 67, 68, 70, 81, 85, 87, 88, 89, 90, 91, 92, 93, 94, 97, 104, 109, 114, 115, 117, 118, 120, 122, 126, 127, 130, 133, 140, 144, 148, 149, 168, 169, 177, 183, 188, or 193 of SEQ ID NO:28.

In some embodiments, the variant CD80 polypeptide has one or more amino acid substitutions selected from the group consisting of:

and conservative amino acid substitutions thereof. In some embodiments, the one or more amino acid substitutions are

In some embodiments, the one or more amino acid substitutions are

In some embodiments, the variant CD80 polypeptide has improved affinity for the CD28 ectodomain or the PD-L1 ectodomain and decreased affinity for the CTLA-4 ectodomain, as compared to a wild-type or unmodified CD80 polypeptide comprising, e.g., a sequence as set forth in SEQ ID NO:28, 152, or 372.

からなる群より選択される1つまたは複数のアミノ酸置換を有する。いくつかの態様において、バリアントCD80ポリペプチドは、

ならびにその保存的アミノ酸置換からなる群より選択される1つまたは複数のアミノ酸置換を有する。いくつかの態様において、1つまたは複数のアミノ酸置換は、

である。いくつかの態様において、バリアントCD80ポリペプチドは、例えばSEQ ID NO:28、152、または372に示される配列を含む野生型または未改変CD80ポリペプチドと比較して、CD28のエクトドメインに対する親和性が向上しており、かつCTLA-4のエクトドメインに対する親和性が低下している。 In some embodiments, the variant CD80 polypeptide has one or more amino acid modifications (e.g., substitutions) corresponding to positions 63, 70, 81, 87, 91, 114, 118, 120, 127, or 130 of SEQ ID NO:28.

In some embodiments, the variant CD80 polypeptide has one or more amino acid substitutions selected from the group consisting of:

and conservative amino acid substitutions thereof. In some embodiments, the one or more amino acid substitutions are

In some embodiments, the variant CD80 polypeptide has improved affinity for the CD28 ectodomain and reduced affinity for the CTLA-4 ectodomain, as compared to a wild-type or unmodified CD80 polypeptide comprising, e.g., a sequence as set forth in SEQ ID NO:28, 152, or 372.

からなる群より選択される1つまたは複数のアミノ酸置換を有する。いくつかの態様において、バリアントCD80ポリペプチドは、

ならびにその保存的アミノ酸置換からなる群より選択される1つまたは複数のアミノ酸置換を有する。いくつかの態様において、1つまたは複数のアミノ酸置換は、

である。いくつかの態様において、1つまたは複数のアミノ酸置換は、

である。いくつかの態様において、バリアントCD80ポリペプチドは、例えばSEQ ID NO:28、152、または372に示される配列を含む野生型または未改変CD80ポリペプチドと比較して、PD-L1のエクトドメインに対する親和性が向上しており、かつCTLA-4のエクトドメインに対する親和性が低下している。 In some embodiments, the variant CD80 polypeptide has one or more amino acid modifications (e.g., substitutions) corresponding to positions 12, 18, 29, 31, 33, 36, 38, 40, 41, 42, 43, 47, 48, 59, 64, 67, 68, 70, 77, 81, 85, 87, 88, 89, 90, 91, 92, 93, 94, 97, 104, 109, 115, 117, 118, 120, 122, 126, 130, 133, 140, 144, 148, 149, 168, 169, 177, 183, 188, or 193 of SEQ ID NO:28.

In some embodiments, the variant CD80 polypeptide has one or more amino acid substitutions selected from the group consisting of:

and conservative amino acid substitutions thereof. In some embodiments, the one or more amino acid substitutions are

In some embodiments, the one or more amino acid substitutions are

In some embodiments, the variant CD80 polypeptide has improved affinity for the ectodomain of PD-L1 and decreased affinity for the ectodomain of CTLA-4, as compared to a wild-type or unmodified CD80 polypeptide comprising, e.g., a sequence as set forth in SEQ ID NO:28, 152, or 372.

In some embodiments, the variant CD80 polypeptide has one or more amino acid modifications (e.g., substitutions) corresponding to positions 70, 81, 87, 91, or 120 of SEQ ID NO:28. In some embodiments, the variant CD80 polypeptide has one or more amino acid substitutions selected from the group consisting of L70Q, Y87N, A91G, and T120S. In some embodiments, the variant CD80 polypeptide has one or more amino acid substitutions selected from the group consisting of L70Q, Y87N, A91G, and T120S, and conservative amino acid substitutions thereof. In some embodiments, the variant CD80 polypeptide has improved affinity for the ectodomain of CD28 and the ectodomain of PD-L1, and reduced affinity for the ectodomain of CTLA-4, as compared to a wild-type or unmodified CD80 polypeptide comprising, for example, a sequence as set forth in SEQ ID NO:28, 152, or 372.

III. Variant Polypeptide Formats
The immunomodulatory polypeptides provided herein, including variant CD80 containing vIgD, can be formatted in a wide variety of ways, for example, as soluble proteins, fusions or conjugates, membrane-bound proteins, secreted proteins, or for delivery in infectious agents. In some embodiments, a particular format can be selected for a desired therapeutic application. In some cases, the immunomodulatory polypeptides, including variant CD80 polypeptides, are provided in a format that antagonizes or blocks the activity of their cognate binding partners (e.g., CD28, PD-L1, and/or CTLA-4). In some embodiments, antagonism against CTLA-4 can be useful for promoting immunity in tumors. In some embodiments, antagonism against CD28 can be useful for treating inflammation or autoimmunity. In some cases, the immunomodulatory polypeptides, including variant CD80 polypeptides, are provided in a format that agonizes or stimulates the activity of their cognate binding partners (e.g., CD28, PD-L1, and/or CTLA-4). In some embodiments, agonism against CTLA-4 can be useful for treating inflammation or autoimmunity. In some embodiments, agonism against CD28 can be useful for treating tumors.Those skilled in the art can easily determine the activity of a particular format, for example, to antagonize or agonize one or more specific cognate binding partners.Exemplary methods for evaluating such activity are provided herein, including in the examples.

In some aspects, immunomodulatory proteins are provided that include the vIgD of CD80, which are soluble, e.g., fused to an Fc chain. In some aspects, one or more additional IgSF domains (e.g., one or more additional vIgDs) can be linked to the vIgD of CD80 as provided herein (hereinafter referred to as "stacked" or "stacked" immunomodulatory proteins). In some embodiments, the modular format of the immunomodulatory proteins provided provides flexibility to modify or generate immunomodulatory proteins to modulate the activity of multiple counter structures (multiple cognate binding partners). In some embodiments, such "stacked" molecules can be provided in a soluble format, and in some cases may be provided as membrane-bound or secreted proteins. In some embodiments, the variant CD80 immunomodulatory protein is provided as a conjugate containing vIgD of CD80 linked directly or indirectly to a targeting agent or moiety (e.g., an antibody or other binding molecule) that specifically binds to a ligand (e.g., an antigen) to, e.g., target or localize vIgD to a particular environment or cell when administered to a subject. In some embodiments, the targeting agent, e.g., an antibody or other binding molecule, binds to a tumor antigen, thereby localizing the variant CD80 containing vIgD to the tumor microenvironment, e.g., modulating the activity of tumor infiltrating lymphocytes (TILs) specific to the tumor microenvironment.

In some embodiments, the provided immunomodulatory proteins are expressed in cells and provided as part of engineered cell therapy (ECT). In some embodiments, the variant CD80 polypeptide is expressed in a cell, e.g., an immune cell (e.g., a T cell or an antigen-presenting cell), in a membrane-bound form, thereby providing a transmembrane immunomodulatory protein (also referred to hereinafter as "TIP"). In some embodiments, depending on the cognate binding partner recognized by TIP, the engineered cell expressing TIP can agonize the cognate binding partner by providing either a positive or negative costimulatory signal to other engineered cells and/or endogenous T cells. In some aspects, the variant CD80 polypeptide is expressed in a secreted form in a cell, e.g., an immune cell (e.g., a T cell or an antigen-presenting cell), thereby producing a secreted or soluble form of the variant CD80 polypeptide (also referred to hereinafter as "SIP"), e.g., when the cell is administered to a subject. In some aspects, the SIP can antagonize its cognate binding partner in the environment in which it is secreted (e.g., the tumor microenvironment). In some embodiments, the variant CD80 polypeptide is expressed in an infectious agent (e.g., a viral or bacterial agent) that is capable of infecting a cell, e.g., an immune cell (e.g., a T cell or an antigen-presenting cell) in vivo upon administration to a subject for delivery or expression of the variant polypeptide into the cell as a TIP or SIP.

In some embodiments, a soluble immunomodulatory polypeptide, such as a variant CD80 containing vIgD, can be encapsulated within a liposome that can itself be conjugated to any one or any combination of the conjugates provided (e.g., targeting moieties). In some embodiments, the soluble or membrane-bound immunomodulatory polypeptides of the invention are deglycosylated. In more specific embodiments, the variant CD80 sequence is deglycosylated. In even more specific embodiments, the IgV and/or IgC (e.g., IgC2) domains of the variant CD80 are deglycosylated.

Non-limiting examples of the formats provided are shown in Figures 1A-1C and further described below.

A. Soluble Proteins In some embodiments, the immunomodulatory protein containing a variant CD80 polypeptide is a soluble protein. Those skilled in the art will recognize that cell surface proteins typically have an intracellular domain, a transmembrane domain, and an extracellular domain (ECD), and that the extracellular domain or its immunologically active sequence can be used to produce a soluble form of such a protein. Thus, in some embodiments, the immunomodulatory protein containing a variant CD80 polypeptide lacks a transmembrane domain or a portion of a transmembrane domain. In some embodiments, the immunomodulatory protein containing a variant CD80 lacks an intracellular (cytoplasmic) domain or a portion of an intracellular domain. In some embodiments, the immunomodulatory protein containing a variant CD80 polypeptide contains only a vIgD portion containing an ECD domain or a portion thereof containing an IgV domain and/or an IgC (e.g., IgC2) domain containing an amino acid modification or a specific binding fragment thereof.

In some embodiments, an immune modulating polypeptide comprising a variant CD80 can comprise one or more variant CD80 polypeptides of the invention. In some embodiments, a polypeptide of the invention comprises exactly one, two, three, four, or five variant CD80 sequences. In some embodiments, at least two of the variant CD80 sequences are the same variant CD80 sequence.

In some embodiments, the immunomodulatory polypeptides provided include two or more vIgD sequences of CD80. The multiple variant CD80 polypeptides within a polypeptide chain can be identical (i.e., homogeneous) or non-identical (i.e., heterogeneous) variant CD80 sequences to each other. In addition to the single polypeptide chain embodiment, in some embodiments, two, three, four, or more of the polypeptides of the invention may be covalently or non-covalently linked to each other. Thus, monomeric, dimeric, and higher order (e.g., 3, 4, 5, or higher) multimeric proteins are provided herein. In some embodiments, for example, exactly two of the polypeptides of the invention can be covalently or non-covalently linked to each other to form a dimer. In some embodiments, the linkage is via an interchain cysteine disulfide bond. A composition comprising two or more polypeptides of the invention can be a composition of identical or substantially identical polypeptides (e.g., homodimers) or non-identical polypeptides (e.g., heterodimers). A composition having multiple linked polypeptides of the invention can have one or more identical or non-identical variant CD80 polypeptides of the invention in each polypeptide chain, as described above.

In some embodiments, the immunomodulatory protein comprises a variant CD80 polypeptide linked to an immunoglobulin Fc (resulting in an "immunomodulatory Fc fusion", e.g., a "variant CD80-Fc fusion", also referred to as a CD80 vIgD-Fc fusion). In some embodiments, the variant CD80 polypeptide is linked at the N-terminus of the Fc. In some embodiments, the variant CD80 polypeptide is linked at the C-terminus of the Fc. In some embodiments, two or more (same or different) CD80 variant polypeptides are independently linked at the N-terminus and C-terminus.

In some embodiments, the Fc is a murine Fc or a human Fc. In some embodiments, the Fc is a mammalian or human IgG1, IgG2, IgG3, or IgG4 Fc region. In some embodiments, the Fc is derived from an IgG1, e.g., a human IgG1. In some embodiments, the Fc comprises an amino acid sequence set forth in SEQ ID NO:226, or an amino acid sequence exhibiting at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or higher sequence identity to SEQ ID NO:226.

In some embodiments, the Fc region contains additional modifications that alter (e.g., reduce) one or more of its normal functions. In general, the Fc region is involved in effector functions, such as complement-dependent cytotoxicity (CDC) and antibody-dependent cellular cytotoxicity (ADCC), in addition to antigen binding, which is the primary function of immunoglobulins. Additionally, the FcRn sequence present in the Fc region plays a role in regulating IgG levels in serum by extending in vivo half-life through binding to the in vivo FcRn receptor. In some embodiments, such functions can be reduced or modified in the Fc for use with the provided Fc fusion proteins.

In some embodiments, one or more amino acid modifications may be introduced into the Fc region of the CD80-Fc variant fusions provided herein, thereby generating an Fc region variant. In some embodiments, the Fc region variant has reduced effector function. There are many examples of changes or mutations to the Fc sequence that can alter effector function. For example, WO 00/42072, WO 2006019447, WO 2012125850, WO 2015/107026, US 2016/0017041, and Shields et al. J Biol. Chem. 9(2): 6591-6604 (2001) describe exemplary Fc variants with improved or reduced binding to FcR. The contents of those publications are specifically incorporated herein by reference.

In some embodiments, the variant CD80-Fc fusions provided comprise an Fc region that exhibits reduced effector function, making them desirable candidates for applications in which the in vivo half-life of the CD80-Fc variant fusion is important but certain effector functions (e.g., CDC and ADCC) are unnecessary or detrimental. In vitro and/or in vivo cytotoxicity assays can be performed to confirm reduced/depleted CDC and/or ADCC activity. For example, Fc receptor (FcR) binding assays can be performed to confirm that the CD80-Fc variant fusion lacks FcγR binding (and thus likely lacks ADCC activity) but retains FcRn binding ability. NK cells, the primary cells for mediating ADCC, express only FcγRIII, whereas monocytes express FcγRI, FcγRII and FcγRIII. FcR expression on hematopoietic cells is summarized in Table 3 on page 464 of Ravetch and Kinet, Annu. Rev. Immunol. 9:457-492 (1991). Non-limiting examples of in vitro assays to assess ADCC activity of a molecule of interest are described in U.S. Patent No. 5,500,362 (see, e.g., Hellstrom, I. et al. Proc. Nat'l Acad. Sci. USA 83:7059-7063 (1986)) and Hellstrom, I et al., Proc. Nat'l Acad. Sci. USA 82:1499-1502 (1985); U.S. Patent No. 5,821,337 (see, Bruggemann, M. et al., J. Exp. Med. 166:1351-1361 (1987)). Alternatively, non-radioactive assay methods may be employed (see, e.g., ACTI™ non-radioactive cytotoxicity assay for flow cytometry (CellTechnology, Inc. Mountain View, Calif.; and CytoTox 96™ non-radioactive cytotoxicity assay (Promega, Madison, Wis.)). Useful effector cells for such assays include peripheral blood mononuclear cells (PBMC) and natural killer (NK) cells. Alternatively, or additionally, ADCC activity of the molecule of interest may be assessed in vivo, e.g., in an animal model such as that disclosed in Clynes et al. Proc. Nat'l Acad. Sci. USA 95:652-656 (1998). C1q binding assays may also be performed to confirm that the CD80-Fc variant fusion is unable to bind C1q and thus lacks CDC activity. See, e.g., WO 2006/029879 and WO See C1q and C3c binding ELISA in 2005/100402. To assess complement activation, CDC assays may be performed (see, e.g., Gazzano-Santoro et al., J. Immunol. Methods 202:163 (1996); Cragg, M. S. et al., Blood 101:1045-1052 (2003); and Cragg, M. S. and M. J. Glennie, Blood 103:2738-2743 (2004)). FcRn binding and in vivo clearance/half-life determinations can also be performed using methods known in the art (see, e.g., Petkova, S. B. et al., Int'l. Immunol. 18(12):1759-1769 (2006)).

CD80-Fc variant fusions with reduced effector function include those with substitutions at one or more of residues 238, 265, 269, 270, 297, 327, and 329 in the Fc region according to EU numbering (U.S. Patent No. 6,737,056). Such Fc variants include Fc variants with substitutions at two or more of amino acid positions 265, 269, 270, 297, and 327 according to EU numbering, including the so-called "DANA" Fc variant in which residues 265 and 297 are substituted with alanine (U.S. Patent No. 7,332,581).

等の改変；WO 2008/092117に記載されているG236R/L328R、L235G/G236R、N325A/L328R、およびN325LL328Rのような改変；位置233、234、235、および237（EUナンバリングによって表示される）におけるアミノ酸挿入；ならびにWO 2000/042072に記載されている部位における改変を含む。 In some embodiments, the Fc region of the CD80-Fc variant fusion has an Fc region in which any one or more of the amino acids at positions 234, 235, 236, 237, 238, 239, 270, 297, 298, 325, and 329 (as indicated by EU numbering) have been substituted with a different amino acid compared to the native Fc region. Such Fc region modifications are not limited to the above modifications and may include, for example, deglycosylated chains as described in Current Opinion in Biotechnology (2009) 20 (6), 685-691.

and modifications such as G236R/L328R, L235G/G236R, N325A/L328R, and N325LL328R described in WO 2008/092117; amino acid insertions at positions 233, 234, 235, and 237 (as indicated by EU numbering); and modifications at sites described in WO 2000/042072.

Certain Fc variants with improved or reduced binding to FcRs have been described (see, e.g., U.S. Pat. No. 6,737,056; WO 2004/056312, WO 2006019447, and Shields et al., J. Biol. Chem. 9(2): 6591-6604 (2001)).

In some embodiments, CD80-Fc variant fusions are provided that include a variant Fc region that includes one or more amino acid substitutions that extend half-life and/or improve binding to the neonatal Fc receptor (FcRn). Antibodies with extended half-life and improved binding to FcRn are described in US2005/0014934A1 (Hinton et al.) or WO 2015107026. These antibodies include an Fc region having one or more substitutions therein that improve binding of the Fc region to FcRn. Such Fc variants include those having a substitution at one or more of the following Fc region residues according to EU numbering: 238, 256, 265, 272, 286, 303, 305, 307, 311, 312, 317, 340, 356, 360, 362, 376, 378, 380, 382, 413, 424, or 434, e.g., a substitution at residue 434 in the Fc region (U.S. Patent No. 7,371,826).

In some embodiments, the Fc region of the CD80-Fc variant fusion comprises one or more of the amino acid substitutions E356D and M358L. In some embodiments, the Fc region of the CD80-Fc variant fusion comprises one or more of the amino acid substitutions C220S, C226S, C229S. In some embodiments, the Fc region of the CD80 variant fusion comprises one or more of the amino acid substitutions R292C and V302C. See also Duncan & Winter, Nature 322:738-40 (1988); U.S. Patent No. 5,648,260; U.S. Patent No. 5,624,821; and WO 94/29351 for other examples of Fc region variants.

In some embodiments, modifications are made in the Fc region that result in reduced C1q binding and/or complement dependent cytotoxicity (CDC), e.g., as described in U.S. Pat. No. 6,194,551, WO 99/51642, and Idusogie et al., J. Immunol. 164: 4178-4184 (2000).

In some embodiments, a CD80-Fc variant fusion is provided that includes a variant Fc region that includes one or more amino acid modifications, where the variant Fc region is derived from an IgG1, e.g., a human IgG1. In some embodiments, the variant Fc region is derived from the amino acid sequence set forth in SEQ ID NO:226. In some embodiments, the Fc contains at least one amino acid substitution that is N82G according to the numbering of SEQ ID NO:226 (corresponding to N297G according to EU numbering). In some embodiments, the Fc further contains at least one amino acid substitution that is R77C or V87C according to the numbering of SEQ ID NO:226 (corresponding to R292C or V302C according to EU numbering). In some embodiments, the variant Fc region further includes a C5S (corresponding to C220S according to EU numbering) amino acid modification according to the numbering of SEQ ID NO:226. For example, in some embodiments, the variant Fc region comprises the following amino acid modifications: N82G and one or more of the amino acid modifications C5S, R77C, or V87C relative to SEQ ID NO:226.

In some embodiments, a CD80-Fc variant fusion is provided that comprises a variant Fc region, wherein the variant Fc comprises an amino acid sequence set forth in any of SEQ ID NOs: 389, 392-394, and 413, or an amino acid sequence that exhibits at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or higher sequence identity to any of SEQ ID NOs: 389, 392-394, and 413.

In some embodiments, the Fc is derived from IgG2, e.g., human IgG2. In some embodiments, the Fc comprises an amino acid sequence set forth in SEQ ID NO:227, or an amino acid sequence that exhibits at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or higher sequence identity to SEQ ID NO:227.

In some embodiments, the Fc comprises an amino acid sequence as set forth in SEQ ID NO:411, or an amino acid sequence that exhibits at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or higher sequence identity to SEQ ID NO:411. In some embodiments, the IgG4 Fc is a stabilized Fc in which the CH3 domain of human IgG4 is replaced with the CH3 domain of human IgG1 and exhibits inhibited aggregate formation, an antibody in which the CH3 and CH2 domains of human IgG4 are replaced with the CH3 and CH2 domains of human IgG1, respectively, or an antibody in which the arginine at position 409 of the EU index proposed by Kabat et al. is replaced with lysine and exhibits inhibited aggregate formation (see, e.g., U.S. Patent No. 8,911,726). In some embodiments, the Fc is an IgG4 containing the S228P mutation, which has been shown to prevent recombination between therapeutic antibodies and endogenous IgG4 by Fab arm exchange (see, e.g., Labrijin et al. (2009) Nat. Biotechnol., 27 (8) 767-71). In some embodiments, the Fc comprises an amino acid sequence set forth in SEQ ID NO:412, or an amino acid sequence that exhibits at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or higher sequence identity to SEQ ID NO:412.

In some embodiments, the variant CD80 polypeptide is directly linked to the Fc sequence. In some embodiments, the variant CD80 polypeptide is indirectly linked to the Fc sequence, e.g., via a linker. In some embodiments, one or more "peptide linkers" link the variant CD80 polypeptide and the Fc domain. In some embodiments, the peptide linker can be a single amino acid residue or longer in length. In some embodiments, the peptide linker has at least one amino acid residue, but is no longer than 20, 19, 18, 17, 16, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, or 1 amino acid residues in length. In some embodiments, the linker is (in the single letter amino acid code) GGGGS ("4GS") or a multimer of the 4GS linker, e.g., two, three, four, or five repeats of the 4GS linker.

In some embodiments, the variant CD80-Fc fusion protein is a dimer formed by two variant CD80 Fc polypeptides linked to an Fc domain. In some specific embodiments, CD80-Fc variant fusion polypeptides of the same or substantially the same species (allowing for three or fewer N-terminal or C-terminal amino acid sequence differences) are dimerized to produce a homodimer. In some embodiments, the dimer is a homodimer in which the two variant CD80 Fc polypeptides are the same. Alternatively, CD80-Fc variant fusion polypeptides of different species can be dimerized to produce a heterodimer. Thus, in some embodiments, the dimer is a heterodimer in which the two variant CD80 Fc polypeptides are different.

Also provided are nucleic acid molecules encoding variant CD80-Fc fusion proteins. In some embodiments, the nucleic acid molecules encoding the variant CD80-Fc fusion proteins are inserted into a suitable expression vector for production of the Fc fusion protein. The resulting variant CD80-Fc fusion protein can be expressed in a host cell transformed with the expression vector, where assembly between the Fc domains occurs via interchain disulfide bonds formed between the Fc moieties to generate dimers such as bivalent variant CD80-Fc fusion proteins.

The resulting Fc fusion protein can be easily purified by affinity chromatography on protein A or protein G columns. For the generation of heterodimers, an additional step of purification may be necessary. For example, when transforming cells with two nucleic acids encoding different variant CD80 polypeptides, the formation of heterodimers must be achieved biochemically, since the variant CD80 molecules carrying the Fc domain are also expressed as disulfide-linked homodimers. Thus, homodimers can be reduced under conditions that favor the disruption of interchain disulfides, but do not affect the interchain disulfides. In some cases, different variant CD80 Fc monomers are mixed and oxidized in equimolar amounts to form a mixture of homodimers and heterodimers. The components of this mixture are separated by chromatographic techniques. Alternatively, the formation of this type of heterodimer can be biased by genetically modifying and expressing Fc fusion molecules containing variant CD80 polypeptides using the knob-into-hole method described below.

B. Stacked Molecules with Additional IgSF Domains In some embodiments, an immunomodulatory protein can contain any of the variant CD80 polypeptides provided herein linked directly or indirectly to one or more other immunoglobulin superfamily (IgSF) domains ("stacked" immunomodulatory protein constructs, also referred to as "type II" immunomodulatory proteins). In some aspects, this can create unique multi-domain immunomodulatory proteins that provide multi-targeted modulation of the immune synapse by binding to two or more, e.g., three or more, cognate binding partners.

In some embodiments, the immunomodulatory protein comprises a combination ("non-wild-type combination") and/or arrangement ("non-wild-type arrangement" or "non-wild-type permutation") of a variant CD80 domain with one or more other affinity-modified and/or affinity-unmodified IgSF domain sequences of another IgSF family member (e.g., a mammalian IgSF family member) not found in a wild-type IgSF family member. In some embodiments, the immunomodulatory protein contains two, three, four, five or six immunoglobulin superfamily (IgSF) domains, where at least one of the IgSF domains is a variant CD80 IgSF domain (vIgD of CD80) according to the description provided.

In some embodiments, the sequence of the additional IgSF domain can be a modified IgSF domain containing one or more amino acid modifications (e.g., substitutions) compared to a wild-type or unmodified IgSF domain. In some embodiments, the IgSF domain can be affinity unmodified (e.g., wild-type) or affinity modified. In some embodiments, the unmodified or wild-type IgSF domain can be from mouse, rat, cynomolgus monkey, or human origin, or a combination thereof. In some embodiments, the additional IgSF domain can be an IgSF domain of an IgSF family member shown in Table 2. In some embodiments, the additional IgSF domain can be an affinity modified IgSF domain containing one or more amino acid modifications (e.g., substitutions) compared to an IgSF domain contained in an IgSF family member shown in Table 2.

In some embodiments, the additional IgSF domain is an affinity modified or unmodified IgSF domain contained in an IgSF family member of a family selected from the following: the signal-regulatory protein (SIRP) family, the triggering receptor-like expressed on myeloid cells (TREML) family, the carcinoembryonic antigen-related cell adhesion molecule (CEACAM) family, the sialic acid-binding Ig-like lectin (SIGLEC) family, the butyrophilin family, the B7 family, the CD28 family, the V-set and immunoglobulin domain-containing (VSIG) family, the V-set transmembrane domain (VSTM) family, the major histocompatibility complex (MHC) family, the signaling lymphocyte activation molecule (SLAM) family, the leukocyte immunoglobulin-like receptor (LIR), the nectin (Nec) family, the nectin-like (NECL) family, the poliovirus receptor-related (PVR) family, the natural cytotoxicity-inducing receptor (NCR) family, the T cell immunoglobulin and mucin (TIM) family, or the killer cell immunoglobulin-like receptor (KIR) family. In some embodiments, the additional IgSF domains are independently derived from an IgSF protein selected from the group consisting of CD80 (B7-1), CD86 (B7-2), CD274 (PD-L1, B7-H1), PDCD1LG2 (PD-L2, CD273), ICOSLG (B7RP1, CD275, ICOSL, B7-H2), CD276 (B7-H3), VTCN1 (B7-H4), CD28, CTLA4, PDCD1 (PD-1), ICOS, BTLA (CD272), CD4, CD8A (CD8α), CD8B (CD8β), LAG3, HAVCR2 (TIM-3), CEACAM1, TIGIT, PVR (CD155), PVRL2 (CD112), CD226, CD2, CD160, CD200, CD200R1 (CD200R), and NC R3 (NKp30).

The first column of Table 2 provides the name and sometimes some possible aliases for that particular IgSF member. The second column provides the protein identifier in the UniProtKB database, a publicly accessible database via the Internet at uniprot.org, and sometimes the GenBank number. The Universal Protein Resource (UniProt) is a comprehensive resource for protein sequence and annotation data. The UniProt database includes the UniProt Knowledgebase (UniProtKB). UniProt is a collaboration between the European Bioinformatics Institute (EMBL-EBI), the SIB Swiss Institute of Bioinformatics, and the Protein Information Resource (PIR), and is supported primarily by grants from the U.S. National Institutes of Health (NIH). GenBank is the NIH gene sequence database, which collects all publicly available DNA sequences with annotations (Nucleic Acids Research, 2013 Jan;41(D1):D36-42). The third column provides the region in which the indicated IgSF domain is located. The region is specified as the range that encompasses the residues that the domain defines. The third column also indicates the IgSF domain class of the specified IgSF region. The fourth column provides the region in which the indicated additional domains are located (signal peptide, S; extracellular domain, E; transmembrane domain, T; cytoplasmic domain, C). It should be understood that the domain descriptions may vary depending on the method used to identify or classify the domains, and may be identified separately from different sources. The descriptions of the residues corresponding to the domains in Table 2 are merely exemplary and may be several amino acids longer or shorter (e.g., 1, 2, 3, or 4). The fifth column indicates some of the IgSF members listed (i.e., some of their cognate cell surface binding partners).

Table 2. IgSF members of the present disclosure

The number of such affinity unmodified or affinity modified IgSF domains (whether in a non-wild-type combination or arrangement) present in a "stacked" immunomodulatory protein construct is at least two, three, four, or five, and in some embodiments, exactly two, three, four, or five IgSF domains (whereby the determination of the number of affinity modified IgSF domains ignores any non-specific binding split sequences thereof and/or split sequences thereof that are substantially immunologically inactive).

In some embodiments of the stacked immunomodulatory proteins provided herein, the number of IgSF domains is at least two, where the number of affinity-modified IgSF domains and the number of affinity-unmodified IgSF domains are each independently at least 0, 1, 2, 3, 4, 5, or 6. Thus, the number of affinity-modified IgSF domains and the number of affinity-unmodified IgSF domains, respectively, can be exactly or at least 2:0 (affinity-modified:wild type), 0:2, 2:1, 1:2, 2:2, 2:3, 3:2, 2:4, 4:2, 1:1, 1:3, 3:1, 1:4, 4:1, 1:5, or 5:1.

In some embodiments of the stacked immunomodulatory proteins, at least two of the non-affinity modified and/or affinity modified IgSF domains are identical IgSF domains.

In some embodiments, the stacked immunomodulatory proteins provided herein include at least two affinity-modified and/or affinity-unmodified IgSF domains from a single IgSF member but in a non-wild-type arrangement (alternatively, "permutation"). One illustrative example of a non-wild-type arrangement or permutation is an immunomodulatory protein of the invention that includes affinity-modified and/or affinity-unmodified IgSF domain sequences of a non-wild-type series compared to those found in wild-type CD80, where the IgSF domain sequences serve as the source of the variant IgSF domains as provided herein. Thus, in one example, an immunomodulatory protein can include an IgV proximal to the transmembrane domain and an IgC distal to the transmembrane domain, regardless of whether the affinity-unmodified and/or affinity-modified form is present. It is also within the scope of the provided subject matter that both non-wild-type combinations and non-wild-type arrangements of affinity-unmodified and/or affinity-modified IgSF domains are present in the immunomodulatory proteins provided herein.

In some embodiments of the stacked immunomodulatory proteins, the affinity unmodified and/or affinity modified IgSF domains are non-identical (i.e., different) IgSF domains. Non-identical affinity modified IgSF domains specifically bind different cognate binding partners under specific binding conditions and are "non-identical" regardless of whether the wild-type or unmodified IgSF domains to which they are modified were the same. Thus, for example, a non-wild-type combination of at least two non-identical IgSF domains in an immunomodulatory protein can include at least one IgSF domain sequence that is unique and originates from one CD80, and at least one second IgSF domain sequence that is unique and originates from another IgSF family member that is not CD80, where the IgSF domains of the immunomodulatory protein are in their affinity unmodified and/or affinity modified forms. However, in alternative embodiments, the two non-identical IgSF domains originate from the same IgSF domain sequence, but at least one is affinity modified such that it specifically binds to a different cognate binding partner.

In some embodiments, the immunomodulatory proteins provided, in addition to containing a variant CD80 polypeptide, also contain at least two, three, four, five, or six additional immunoglobulin superfamily (IgSF) domains (e.g., an IgD domain of an IgSF family member shown in Table 2). In some embodiments, the immunomodulatory proteins provided contain at least one additional IgSF domain (e.g., a second IgSF domain), which is an IgSF domain set forth in a wild-type or unmodified IgSF domain contained in an amino acid sequence set forth in any of SEQ ID NOs:1-27 and 348, or a specific binding fragment thereof. In some embodiments, the wild-type or unmodified IgSF domain is an IgV domain or an IgC domain, e.g., an IgC1 or IgC2 domain.

In some embodiments, the provided immunomodulatory proteins, in addition to containing a variant CD80 polypeptide, also contain at least one additional IgSF domain (e.g., a second IgSF domain) that is a vIgD that contains one or more amino acid modifications (e.g., substitutions, deletions, or mutations) relative to an IgSF domain in a wild-type or unmodified IgSF domain, e.g., an IgSF domain in an IgSF family member set forth in Table 2. In some embodiments, the additional or second affinity-modified IgSF domain comprises at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or higher sequence identity to a wild-type or unmodified IgSF domain, or a specific binding fragment thereof, contained in an amino acid sequence set forth in any of SEQ ID NOs: 1-27 and 348. In some embodiments, the wild-type or unmodified IgSF domain is an IgV domain or an IgC domain, e.g., an IgC1 or IgC2 domain. In some embodiments, the additional or second IgSF domain is an affinity-modified IgV domain or an IgC domain.

In some embodiments, the one or more additional IgSF domains (e.g., a second IgSF domain) are an IgSF domain of another IgSF family member (e.g., IgV) that binds to or recognizes a tumor antigen. In such embodiments, the IgSF family member serves as a tumor-localizing moiety, thereby bringing the vIgD of CD80 into close proximity to immune cells within the tumor microenvironment. In some embodiments, the additional IgSF domain (e.g., a second IgSF domain) is an IgSF domain of NkP30 that binds to or recognizes B7-H6 expressed on tumor cells. In some embodiments, at least one additional (e.g., a second) IgSF domain, e.g., NkP30, is a vIgD that contains one or more amino acid modifications (e.g., substitutions, deletions, or additions). In some embodiments, the one or more amino acid modifications improve binding affinity and/or selectivity for B7-H6, e.g., at least or at least about 1.2-fold, 1.5-fold, 2-fold, 3-fold, 4-fold, 5-fold, 6-fold, 7-fold, 8-fold, 9-fold, 10-fold, 20-fold, 30-fold, 40-fold, or 50-fold, as compared to an unmodified IgSF domain (e.g., NkP30). Among the exemplary polypeptides is an NKp30 variant containing the mutations L30V/A60V/S64P/S86G relative to the positions of the NKp30 extracellular domain corresponding to the positions shown in SEQ ID NO:54.

Tables 3-5 provide exemplary polypeptides containing one or more affinity-modified IgSF domains that can be used in the stack constructs provided herein.

Table 3: Exemplary variant ICOSL polypeptides

Table 4. Exemplary variant NKp30 polypeptides

Table 5. Exemplary variant CD86 polypeptides

In some embodiments, two or more IgSF domains, including the vIgD of CD80 and one or more additional IgSF domains from another IgSF family member (e.g., a second variant IgSF domain), are covalently or non-covalently linked. The multiple affinity unmodified and/or affinity modified IgSF domains in a stacked immunomodulatory protein polypeptide chain need not be directly covalently linked to each other. In some embodiments, two or more IgSF domains are directly or indirectly linked, for example, via a linker. In some embodiments, an intervening span of one or more amino acid residues indirectly covalently links the IgSF domains to each other. Linkage can be through residues from the N-terminus to the C-terminus. In some embodiments, linkage can be through a side chain of an amino acid residue that is not located at the N-terminus or C-terminus of the IgSF domain. Thus, linkage can be through terminal or internal amino acid residues or a combination thereof.

In some embodiments, one or more "peptide linkers" link the vIgD of CD80 to the additional IgSF domain (e.g., a second variant IgSF domain). In some embodiments, the peptide linker can be a single amino acid residue or longer in length. In some embodiments, the peptide linker has at least one amino acid residue, but is no longer than 20, 19, 18, 17, 16, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, or 1 amino acid residues in length. In some embodiments, the linker is (in the single letter amino acid code): GGGGS ("4GS") or a multimer of 4GS linkers, such as 2, 3, 4, or 5 repeats of the 4GS linker. In some embodiments, the peptide linker is (GGGGS) ₂ or (GGGGS) ₃ . In some embodiments, the linker can also include a series of alanine residues, alone or in addition to another peptide linker (e.g., a 4GS linker or multimer thereof). In some embodiments, the number of alanine residues in each series is 2, 3, 4, 5, or 6 alanines.

In some embodiments, the affinity unmodified and/or affinity modified IgSF domains are linked by a "wild-type peptide linker" inserted at the N-terminus and/or C-terminus of the first and/or second affinity unmodified and/or affinity modified IgSF domains. These linkers are also referred to as leading sequences (N-terminus of the affinity unmodified or affinity modified IgSF domain) or trailing sequences (C-terminus of the affinity unmodified or affinity modified IgSF domain), and sequences present in the wild type that extend just outside the structural prediction of the Ig fold of IgSF. In some embodiments, a "wild-type linker" is an amino acid sequence that is present in the amino acid sequence of the wild-type protein after the signal sequence but before the IgSF domain (e.g., a defined IgV domain). In some embodiments, a "wild-type" linker is an amino acid sequence that is present in the amino acid sequence of the wild-type protein immediately after the IgSF domain (e.g., immediately after the defined IgV domain) but before the IgC domain. These linker sequences can contribute to the proper folding and function of adjacent IgSF domains. In some embodiments, there is a leading peptide linker inserted at the N-terminus of the first IgSF domain and/or a trailing sequence inserted at the C-terminus of the first affinity-unmodified and/or affinity-modified IgSF domain. In some embodiments, there is a second leading peptide linker inserted at the N-terminus of the second IgSF domain and/or a second trailing sequence inserted at the C-terminus of the second affinity-unmodified and/or affinity-modified IgSF domain. When the first and second affinity-unmodified and/or affinity-modified IgSF domains are derived from the same parent protein and are connected in the same orientation, the wild-type peptide linkers between the first and second affinity-unmodified and/or affinity-modified IgSF domains do not overlap. For example, when the first trailing wild-type peptide linker and the second leading wild-type peptide linker are the same, the type II immunomodulatory protein does not include either the first trailing wild-type peptide linker or the second leading wild-type peptide linker.

In some embodiments, the type II immunomodulatory protein comprises a first leading wild-type peptide linker inserted at the N-terminus of the first affinity unmodified and/or affinity modified IgSF domain, wherein the first leading wild-type peptide linker comprises at least 5 (e.g., at least about any of 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 or more) contiguous amino acids from an intervening sequence in the wild-type protein from which the first affinity unmodified and/or affinity modified IgSF domain is derived, between the parent IgSF domain and the immediately preceding domain (e.g., a signal peptide or IgSF domain). In some embodiments, the first leading wild-type peptide linker comprises the entire intervening sequence in the wild-type protein from which the first affinity unmodified and/or affinity modified IgSF domain is derived, between the parent IgSF domain and the immediately preceding domain (e.g., a signal peptide or IgSF domain).

In some embodiments, the type II immunomodulatory protein further comprises a first trailing wild-type peptide linker inserted at the C-terminus of the first affinity unmodified and/or affinity modified IgSF domain, wherein the first trailing wild-type peptide linker comprises at least 5 (e.g., at least about any of 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 or more) contiguous amino acids from an intervening sequence in the wild-type protein from which the first affinity unmodified and/or affinity modified IgSF domain is derived, between the parent IgSF domain and the immediately following domain (e.g., IgSF domain or transmembrane domain). In some embodiments, the first trailing wild-type peptide linker comprises the entire intervening sequence in the wild-type protein from which the first affinity unmodified and/or affinity modified IgSF domain is derived, between the parent IgSF domain and the immediately following domain (e.g., IgSF domain or transmembrane domain).

In some embodiments, the type II immunomodulatory protein further comprises a second leading wild-type peptide linker inserted at the N-terminus of the second affinity unmodified and/or affinity modified IgSF domain, wherein the second leading wild-type peptide linker comprises at least 5 (e.g., at least about any of 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 or more) contiguous amino acids from an intervening sequence in the wild-type protein from which the second affinity unmodified and/or affinity modified IgSF domain is derived, between the parent IgSF domain and the immediately preceding domain (e.g., a signal peptide or IgSF domain). In some embodiments, the second leading wild-type peptide linker comprises the entire intervening sequence in the wild-type protein from which the second affinity unmodified and/or affinity modified IgSF domain is derived, between the parent IgSF domain and the immediately preceding domain (e.g., a signal peptide or IgSF domain).

In some embodiments, the type II immunomodulatory protein further comprises a second trailing wild-type peptide linker inserted at the C-terminus of the second affinity unmodified and/or affinity modified IgSF domain, wherein the second trailing wild-type peptide linker comprises at least 5 (e.g., at least about any of 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 or more) contiguous amino acids from an intervening sequence in the wild-type protein from which the second affinity unmodified and/or affinity modified IgSF domain is derived, between the parent IgSF domain and the immediately following domain (e.g., IgSF domain or transmembrane domain). In some embodiments, the second trailing wild-type peptide linker comprises the entire intervening sequence in the wild-type protein from which the second affinity unmodified and/or affinity modified IgSF domain is derived, between the parent IgSF domain and the immediately following domain (e.g., IgSF domain or transmembrane domain).

An exemplary trailing sequence for a type II protein containing a CD80 IgSF domain may contain the amino acid sequence shown in SEQ ID NO:232. Exemplary trailing sequences for a type II protein containing a CD80 IgSF domain are shown in SEQ ID NOs:231, 232, and 371. Exemplary trailing sequences for a type II protein containing an ICOSL IgSF domain are shown in SEQ ID NOs:233 and 234. Exemplary leading and trailing sequences for a type II protein containing a CD86 IgSF domain are shown in SEQ ID NOs:236-238. An exemplary trailing sequence for a type II protein containing an NKp30 IgSF domain is shown in SEQ ID NO:235.

In some embodiments, two or more IgSF domains, including the vIgD of CD80 and one or more additional IgSF domains from another IgSF family member (e.g., a second variant IgSF domain), are linked or attached to Fc to form a dimeric multidomain stacked immunomodulatory protein. In some embodiments, the variant CD80 polypeptide and the second IgSF domain are independently linked, directly or indirectly, to the N-terminus or C-terminus of the Fc subunit. In some embodiments, the variant CD80 polypeptide and the second IgSF domain are directly or indirectly linked, and one of the variant CD80 or the second IgSF domain is also directly or indirectly linked to the N-terminus or C-terminus of the Fc subunit. In some embodiments, the link to the Fc is via a peptide linker (e.g., a peptide linker as described above). In some embodiments, the link between the variant CD80 and the second IgSF domain is via a peptide linker (e.g., a peptide linker as described above). In some embodiments, the vIgD of CD80, one or more additional IgSF domains, and the Fc domain can be linked together in any of a number of configurations as shown in FIG. 4. Exemplary configurations are described in the Examples.

In some embodiments, the stacked immunomodulatory protein is a dimer formed by two stacked immunomodulatory Fc fusion polypeptides. Nucleic acid molecules encoding any of the stacked immunomodulatory proteins are also provided. In some embodiments, the dimeric multidomain stacked immunomodulatory protein can be produced in a cell by expression, or in some cases, co-expression, of, for example, the stacked immunomodulatory Fc fusion polypeptides described above, following the generation of the dimeric Fc fusion protein.

In some embodiments, the dimeric multidomain stacked immunomodulatory protein is bivalent for each Fc subunit, monovalent for each subunit, or bivalent for one subunit and tetravalent for the other.

In some embodiments, the dimeric multidomain stacked immunomodulatory protein is a homodimeric multidomain stacked Fc protein. In some embodiments, the dimeric multidomain stacked immunomodulatory protein comprises a first stacked immunomodulatory Fc fusion polypeptide and a second stacked immunomodulatory Fc fusion polypeptide, the first and second polypeptides being the same. In some embodiments, the Fc portion of the polypeptide can be any Fc, as described above.

In some embodiments, the multidomain stack molecule is a heterodimer comprising two different Fc polypeptides, at least one of which is an Fc polypeptide containing at least one variant CD80 polypeptide and/or at least one second IgSF domain (e.g., a second variant IgSF domain). In some embodiments, the multidomain stack molecule contains a first Fc polypeptide containing a variant CD80 and a second IgSF domain and a second Fc polypeptide containing a variant CD80 and a second IgSF domain. In some embodiments, the multidomain stack molecule contains a first Fc polypeptide containing a variant CD80 polypeptide and a second IgSF domain and a second Fc polypeptide that is not linked to either the variant CD80 polypeptide or the second IgSF domain.

In some embodiments, the multidomain stack molecule contains a first Fc polypeptide containing one, two, three, four, or more variant CD80 polypeptides and one, two, three, four, or more second IgSF domains, such that the total number of IgSF domains in the first stack Fc polypeptide is two, three, four, five, six, or more. In one such embodiment, the second stack Fc polypeptide contains one, two, three, four, or more variant CD80 polypeptides and one, two, three, four, or more second IgSF domains, such that the total number of IgSF domains in the first stack Fc polypeptide is two, three, four, five, six, or more. In another such embodiment, the second Fc polypeptide is not linked to either the variant CD80 polypeptide or the second IgSF domain.

In some embodiments, the heterodimeric stack molecule contains a first stacked immunomodulatory Fc fusion polypeptide and a second stacked immunomodulatory Fc fusion polypeptide, the first and second polypeptides being different. In some embodiments, the heterodimeric stack molecule contains a first Fc subunit containing a first variant CD80 polypeptide and/or a second IgSF domain (e.g., a second variant IgSF domain), and a second Fc subunit containing the first variant CD80 polypeptide or another of the second IgSF domain. In some embodiments, the heterodimeric stack molecule contains a first stacked immunomodulatory Fc fusion polypeptide and a second stacked immunomodulatory Fc fusion polypeptide, the first and second polypeptides being different. In some embodiments, the heterodimeric stack molecule contains a first Fc subunit that contains a first variant CD80 polypeptide and/or a second IgSF domain (e.g., a second variant IgSF domain) and a second Fc subunit that contains both the first variant CD80 polypeptide and a second IgSF domain (e.g., a second variant IgSF domain) but in a different orientation or configuration than the first Fc subunit.

In some embodiments, the Fc domain of one or both of the first and second stacked immunomodulatory Fc fusion polypeptides includes modifications (e.g., substitutions) such that the interface of the Fc molecules is modified to facilitate and/or promote heterodimerization. In some embodiments, the modifications include the introduction of a knob into the first Fc polypeptide and a hole into the second Fc polypeptide, which allows the knob to be positioned within the cavity to facilitate complexation of the first and second Fc-containing polypeptides. The amino acids targeted for replacement and/or modification to create the knob or cavity within the polypeptide are typically interface amino acids that interact or contact one or more amino acids at the interface of the second polypeptide.

In some embodiments, the first polypeptide modified to contain a protrusion (hole) amino acid comprises a replacement of a natural or original amino acid with an amino acid having at least one side chain that protrudes from the interface of the first polypeptide and is therefore capable of positioning in a compensatory cavity (hole) in the adjacent interface of the second polypeptide. In most cases, the replacement amino acid has a larger side chain volume than the original amino acid residue. Those skilled in the art will know how to determine and/or evaluate the properties of an amino acid residue to identify an ideal replacement amino acid that creates a protrusion. In some embodiments, the replacement residue for the formation of the protrusion is a naturally occurring amino acid residue, including, for example, arginine (R), phenylalanine (F), tyrosine (Y), or tryptophan (W). In some examples, the original residue identified for replacement is an amino acid residue with a small side chain, such as alanine, asparagine, aspartic acid, glycine, serine, threonine, or valine.

In some embodiments, the second polypeptide modified to contain a cavity (hole) comprises a replacement of a natural or original amino acid with an amino acid having at least one side chain that is recessed from the interface of the second polypeptide and can therefore accommodate a corresponding protrusion from the interface of the first polypeptide. In most cases, the replacement amino acid has a smaller side chain volume than the original amino acid residue. Those skilled in the art know how to determine and/or evaluate the properties of amino acid residues to identify those that are ideal replacement residues for the formation of a cavity. Generally, replacement residues for the formation of a cavity are naturally occurring amino acids, including, for example, alanine (A), serine (S), threonine (T) and valine (V). In some examples, the original amino acid identified for replacement is an amino acid with a large side chain, for example, tyrosine, arginine, phenylalanine, or tryptophan.

The CH3 interface of human IgG1, for example, involves 16 residues on each domain located on four antiparallel β-strands buried 1090 Å from each surface (see, e.g., Deisenhofer et al. (1981) Biochemistry, 20:2361-2370; Miller et al., (1990) J Mol. Biol., 216, 965-973; Ridgway et al., (1996) Prot. Engin., 9: 617-621; U.S. Pat. No. 5,731,168). Modifications of the CH3 domain to create protrusions or cavities have been described, for example, in U.S. Patent No. 5,731,168; International Patent Applications WO 98/50431 and WO 2005/063816; and Ridgway et al., (1996) Prot. Engin., 9: 617-621. In some instances, modifications of the CH3 domain to create protrusions or cavities are typically targeted to residues located on the two central antiparallel β-strands. The aim is to minimize the risk that the created protrusions may be accommodated by protruding into the surrounding solvent rather than being accommodated by a compensatory cavity in the partner CH3 domain.

In some embodiments, the heterodimeric molecule contains a T366W mutation in the CH3 domain of the "knob chain" and a T366S, L368A, Y407V mutation in the CH3 domain of the "hole chain". In some cases, additional interchain disulfide bridges between the CH3 domains (Merchant, A. M., et al., Nature Biotech. 16 (1998) 677-681) can also be used, for example, by introducing a Y349C mutation in the CH3 domain of the "knob" or "hole" chain and an E356C or S354C mutation in the CH3 domain of the other chain. In some embodiments, the heterodimeric molecule contains an S354C, T366W mutation in one of the two CH3 domains and a Y349C, T366S, L368A, Y407V mutation in the other of the two CH3 domains. In some embodiments, the heterodimeric molecule comprises an E356C, T366W mutation in one of the two CH3 domains and an Y349C, T366S, L368A, Y407V mutation in the other of the two CH3 domains. In some embodiments, the heterodimeric molecule comprises an Y349C, T366W mutation in one of the two CH3 domains and an E356C, T366S, L368A, Y407V mutation in the other of the two CH3 domains. In some embodiments, the heterodimeric molecule comprises an Y349C, T366W mutation in one of the two CH3 domains and an S354C, T366S, L368A, Y407V mutation in the other of the two CH3 domains. Other examples of knob-in-hole technology are known in the art, for example, as described in EP 1 870 459 A1.

In some embodiments, the Fc subunit of the heterodimeric molecule can additionally contain one or more other Fc mutations, such as any of those described above. In some embodiments, the heterodimeric molecule contains an Fc subunit with a mutation that reduces effector function.

In some embodiments, the Fc variant containing a CH3 protrusion/cavity modification can be connected anywhere in the stacked immunomodulatory polypeptide, but typically via its N-terminus or C-terminus, to the N-terminus or C-terminus of the first and/or second stacked immunomodulatory polypeptide, for example to form a fusion polypeptide. The linkage can be direct or indirect via a linker. Typically, the knob and hole molecules are generated by co-expression of a first stacked immunomodulatory polypeptide linked to an Fc variant containing a CH3 protrusion modification and a second stacked immunomodulatory polypeptide linked to an Fc variant containing a CH3 cavity modification.

C. Conjugates and Fusions of Variant Polypeptides and Immunomodulatory Proteins In some embodiments, the variant polypeptides provided herein, which are immunomodulatory proteins comprising variants of the IgSF family of Ig domains (vIgD), can be conjugated or fused, directly or indirectly, to a moiety, such as an effector moiety, such as another protein, to form a conjugate ("IgSF conjugate"). In some embodiments, the binding can be covalent or non-covalent (e.g., via biotin-streptavidin non-covalent interactions). In some embodiments of the CD80-Fc variant fusion, any one or combination of any two or more of the aforementioned conjugates can be attached to the Fc or variant CD80 polypeptide or both.

In some embodiments, the moiety can be a targeting moiety, a small molecule drug (a non-polypeptide drug with a molar mass of less than 500 Daltons), a toxin, a cytostatic agent, a cytotoxic agent, an immunosuppressant, a radioactive substance suitable for diagnostic purposes, a radioactive metal ion for therapeutic purposes, a prodrug activating enzyme, a substance that extends biological half-life, or a diagnostic or detectable substance.

In some embodiments, the effector moiety is a therapeutic agent, e.g., a cancer therapeutic agent, that provides cytotoxicity, cytostaticity, or otherwise some therapeutic benefit. In some embodiments, the effector moiety is a targeting moiety or agent, e.g., an agent that targets a cell surface antigen, e.g., an antigen on the surface of a tumor cell. In some embodiments, the effector moiety is a label that can generate a detectable signal, either directly or indirectly. In some embodiments, the effector moiety is a toxin. In some embodiments, the effector moiety is a protein, peptide, nucleic acid, small molecule, or nanoparticle.

In some embodiments, one, two, three, four, five or more effector moieties, which may be the same or different, are conjugated, linked or fused to the variant polypeptide or protein to form an IgSF conjugate. In some embodiments, such effector moieties can be attached to the variant polypeptide or immunomodulatory protein using various molecular biological or chemical conjugation and linking methods known in the art and described below. In some embodiments, the effector moieties can be linked or conjugated to the variant polypeptide or immunomodulatory protein using a linker, such as a peptide linker, a cleavable linker, a non-cleavable linker, or a linker that aids in the conjugation reaction.

In some embodiments, the IgSF conjugate comprises the following components: (protein or polypeptide), (L) _q and (effector moiety) _m , where the protein or polypeptide is any of the described variant polypeptides or immunomodulatory proteins capable of binding to one or more cognate counterstructure ligands as described; L is a linker for linking the protein or polypeptide to the moiety; m is at least 1; q is 0 or greater; and the resulting IgSF conjugate binds to one or more counterstructure ligands. In certain embodiments, m is 1-4 and q is 0-8.

In some embodiments, an IgSF conjugate is provided that includes a variant polypeptide or an immunomodulatory protein provided herein conjugated to a targeting agent that binds to a cell surface molecule, e.g., for targeted delivery of the variant polypeptide or immunomodulatory protein to a specific cell. In some embodiments, the targeting agent is a molecule that has the ability to localize and bind to a molecule present on normal cells/tissues and/or tumor cells/tumors in a subject. In other words, an IgSF conjugate that includes a targeting agent can bind (directly or indirectly) to a ligand present on a cell, e.g., a tumor cell. Targeting agents of the present invention contemplated for use include antibodies, polypeptides, peptides, aptamers, other ligands, or any combination thereof, that can bind to a component of a target cell or molecule.

In some embodiments, the targeting agent can bind to or near tumor cells (e.g., tumor vasculature or tumor microenvironment) after administration to a subject. The targeting agent can bind to a receptor or ligand on the surface of a cancer cell. In another aspect of the invention, a targeting agent is selected that is specific for a non-cancerous cell or tissue. For example, the targeting agent can be specific for a molecule that is normally present on a particular cell or tissue. Furthermore, in some embodiments, the same molecule can be present on normal cells and cancer cells. A variety of cellular components and molecules are known. For example, if the targeting agent is specific for EGFR, the resulting IgSF conjugate can target both cancer cells that express EGFR as well as normal skin epithelial cells that express EGFR. Thus, in some embodiments, the IgSF conjugates of the invention can work by two separate mechanisms (targeting cancer cells and non-cancerous cells).

In various aspects of the invention disclosed herein, the IgSF conjugates of the invention comprise a targeting agent capable of binding/targeting a cellular component, e.g., a tumor antigen, a bacterial antigen, a viral antigen, a mycoplasma antigen, a fungal antigen, a prion antigen, an antigen from a parasite. In some aspects, the cellular component, antigen, or molecule can each be used to refer to the desired target of the targeting agent. For example, in various embodiments, the targeting agent is selected from the group consisting of epidermal growth factor receptor (EGFR, ErbB-1, HERl), ErbB-2 (HER2/neu), ErbB-3/HER3, ErbB-4/HER4, EGFR ligand family; insulin-like growth factor receptor (IGFR) family, IGF binding protein (IGFBP), IGFR ligand family; platelet-derived growth factor receptor (PDGFR) family, PDGFR ligand family; fibroblast growth factor receptor (FGFR) family, FGFR ligand family, vascular endothelial cell growth factor (VEGF) family, VEGF-binding protein (IGFBP), ... long-acting factor receptor (VEGFR) family, VEGF family; HGF receptor family; TRK receptor family; ephrin (EPH) receptor family; AXL receptor family; leukocyte tyrosine kinase (LTK) receptor family; TIE receptor family, angiopoietin 1,2; receptor tyrosine kinase-like orphan receptor (ROR) receptor family, e.g., ROR1; CD171 (L1CAM); B7-H6 (NCR3LG1); PD-L1, tumor glycosylated antigens, e.g., sTn or Tn, e.g., sTn of MUC1 Ag; LHR (LHCGR); phosphatidylserine, discoidin domain receptor (DDR) family; RET receptor family; KLG receptor family; RYK receptor family; MuSK receptor family; transforming growth factor-α (TGF-α) receptor, TGF-β; cytokine receptors, class I (hematopoietin family) and class II (interferon/IL-10 family) receptors, tumor necrosis factor (TNF) receptor superfamily (TNFRSF), death receptor family; cancer-testis (CT) antigens, lineage-specific antigens, differentiation antigens, alpha-actinin-4, ARTCl, breakpoint cluster region-Abelson (Bcr-abl) fusion products, B-RAF, caspase-5 (CASP-5), caspase-8 (CASP-8), β-catenin (CTNNBl), cell division cycle 27 (CDC27), cyclin-dependent kinase 4 (CDK4), CDKN2A, COA-I, dek-can fusion protein, EFTUD-2, elongation factor 2 (ELF2), Ets variant gene 6/acute myeloid leukemia 1 gene ETS (ETC6-AML1) fusion protein, fibronectin (FN), e.g., fibronectin ectodomain A (EDA), GPNMB, low-density lipid receptor/GDP-L fucose:β-D galactose 2-α-L fucosyltransferase (LDLR/FUT) fusion protein, HLA-A2. arginine-isoleucine exchange at residue 170 of the α-helix of the α2 domain in the HLA-A2 gene (HLA-A*201-R170I), HLA-Alpha-A2. 1, heat shock protein 70-2 mutant (HSP70-2M), K1AA0205, MART2, melanoma ubiquitous mutant 1, 2, 3 (MUM-I, 2, 3), prostatic acid phosphatase (PAP), neo-PAP, myosin class I, NFYC, OGT, OS-9, pml-RAR alpha fusion protein, PRDX5, PTPRK, K-ras (KRAS2), N-ras (NRAS), HRAS, RBAF600, SIRT2, SNRPDl, SYT-SSXl or -SSX2 fusion protein, triose phosphate isoform merase, BAGE, BAGK-1, BAGE-2,3,4,5, GAGE-1,2,3,4,5,6,7,8, GnT-V (abnormal N-acetylglucosaminyltransferase V, MGAT5), HERV-K-MEL, KK-LC, KM-HN-I, LAGE, LAGE-I, CTL-recognized antigen on melanoma (CAMEL), MAGE-Al (MAGE-I), MAGE-A2, MAGE-A3, MAGE-A4, MAGE-A5, MAGE-A6, MAGE-A8, MAGE-A9, MAGE-AlO, MAGE-AI l, MAGE-A12, MAGE-3, MAGE-Bl, MAGE-B2, MAGE-B5, MAGE-B6, MAGE-Cl, MAGE-C2, Mucin 1 (MUCl), MART-1/Melan-A (MLANA), gplOO, gplOO/Pmell7 (SILV), Tyrosinase (TYR), TRP-I, HAGE, NA-88, NY-ESO-I, NY-ESO-l/LAGE-2, SAGE, Spl 7, SSX-1,2,3,4, TRP2-INT2, carcinoembryonic antigen (CEA), kallikrein 4, mammaglobin-A, OAl, prostate-specific antigen (PSA), TRP-1/gp75, TRP-2, adipophilin, interferon-inducible protein 2 (AIM-2) absent in melanoma, BING-4, CPSF, cyclin Dl, epithelial cell adhesion molecule (Ep-CAM), EphA3, fibroblast growth factor-5 (FGF-5), glycoprotein 250 (gp250), EGFR (ERBBl), HER-2/neu (ERBB2), interleukin-13 receptor alpha 2 chain (IL13Rα2), IL-6 receptor, intestinal carboxylesterase (iCE), alpha-fetoprotein (AFP), M-CSF, mdm-2, MUCl, p53 (TP53), PBF, PRAME, PSMA, RAGE-I, RNF43, RU2AS, SOXlO, STEAPl, survivin (BIRC5), human telomerase reverse transcriptase (hTERT), telomerase, Wilms tumor gene (WTl), SYCPl, BRDT, SPANX, XAGE, ADAM2, PAGE-5, LIPl, CTAGE-I, CSAGE, MMAl, CAGE, BORIS, HOM-TES-85, AF15ql4, HCA661, LDHC, MORC, SGY-I, SPOl 1, TPXl, NY-SAR-35, FTHL17, NXF2, TDRDl, TEX15, FATE, TPTE, immunoglobulin idiotype, Bence-Jones protein, estrogen receptor (ER), androgen receptor (AR), CD40, CD30, CD20, CD19, CD33, cancer antigen 72-4 (CA 72-4), cancer antigen 15-3 (CA 15-3), cancer antigen 27-29 (CA 27-29), cancer antigen 125 (CA 125), cancer antigen 19-9 (CA 19-9), beta-human chorionic gonadotropin, beta-2 microglobulin, squamous cell carcinoma antigen, neurospecific enolase, heat shock protein gp96, GM2, sargramostim, CTLA-4, 707 alanine proline (707-AP), adenocarcinoma antigen recognized by T cells 4 (ART-4), carcinoembryonic antigen peptide-1 (CAP-I), calcium activated chloride channel-2 (CLCA2), cyclophilin B (Cyp-B), human signet ring tumor-2 (HST-2), human papillomavirus (HPV) proteins (HPV-E6, HPV-E7, major or minor capsid antigen, etc.), Epstein-Barr virus (EBV) proteins (EBV latent membrane protein-LMP1, LMP2; etc.), hepatitis B or C virus proteins, and HIV proteins.

In some embodiments, the IgSF conjugate, by virtue of its targeting agent, binds to tumor cells, tumor vasculature, or cellular components of the tumor microenvironment, thereby promoting killing of the targeted cells through modulation of the immune response (e.g., by activation of costimulatory molecules or inhibition of negative regulatory molecules of immune cell activation), inhibition of survival signals (e.g., growth factor or cytokine or hormone receptor antagonists), activation of death signals, and/or immune-mediated cytotoxicity (e.g., by antibody-dependent cellular cytotoxicity). Such IgSF conjugates can function by several mechanisms to suppress, reduce, or eliminate tumor cells, e.g., to facilitate delivery of conjugated effector moieties to tumor targets, e.g., by receptor-mediated endocytosis of the IgSF conjugate; or such conjugates can recruit, bind, and/or activate immune cells (e.g., NK cells, monocytes/macrophages, dendritic cells, T cells, B cells). Additionally, in some cases, one or more of the aforementioned pathways may be acted upon by administration of one or more IgSF conjugates of the present invention.

In some embodiments, the IgSF conjugate, through its targeting agent, localizes, e.g., binds, to tumor cells, tumor vasculature, or cellular components of the tumor microenvironment, thereby modulating cells of the immune response in the vicinity of the tumor. In some embodiments, the targeting agent facilitates delivery of the conjugated IgSF (e.g., vIgD) to the tumor target, e.g., interacts with its cognate binding partner, and alters signaling of immune cells (e.g., NK cells, monocytes/macrophages, dendritic cells, T cells, B cells) that bear the cognate binding partner. In some embodiments, the localized delivery mediates costimulatory signaling via CD28 or ICOS.

In some embodiments, the targeting agent is an immunoglobulin. As used herein, the term "immunoglobulin" includes natural or artificial monovalent or polyvalent antibodies, including, but not limited to, polyclonal, monoclonal, multispecific, human, humanized or chimeric antibodies, single chain antibodies, Fab fragments, F(ab') fragments, fragments produced by a Fab expression library, single chain Fvs (scFvs); anti-idiotypic (anti-Id) antibodies (including, for example, anti-Id antibodies to the antibodies of the invention), and epitope-binding fragments of any of the above. The term "antibody" as used herein refers to immunoglobulin molecules and immunologically active portions of immunoglobulin molecules, e.g., molecules that contain an antigen-binding site that immunospecifically binds to an antigen. The immunoglobulin molecules of the invention can be of any type (e.g., IgG, IgE, IgM, IgD, IgA, and IgY), class (e.g., IgGl, IgG2, IgG3, IgG4, IgAl, and IgA2) or subclass of immunoglobulin molecule.

In some embodiments, the IgSF conjugate, through its antibody targeting moiety, binds to tumor cells, tumor vasculature, or cellular components of the tumor microenvironment, thereby promoting apoptosis of the targeted cells through modulation of the immune response (e.g., by activating costimulatory molecules or inhibiting negative regulatory molecules of immune cell activation), inhibition of survival signals (e.g., growth factor or cytokine or hormone receptor antagonists), activation of death signals, and/or immune-mediated cytotoxicity (e.g., by antibody-dependent cellular cytotoxicity). Such IgSF conjugates can function through several mechanisms to suppress, reduce, or eliminate tumor cells, e.g., to facilitate delivery of conjugated effector moieties to tumor targets, e.g., through receptor-mediated endocytosis of the IgSF conjugate; or such conjugates can recruit, bind, and/or activate immune cells (e.g., NK cells, monocytes/macrophages, dendritic cells, T cells, B cells).

In some embodiments, the IgSF conjugate, through its antibody targeting moiety, binds to tumor cells, tumor vasculature, or cellular components of the tumor microenvironment, thereby modulating the immune response (e.g., by activating costimulatory molecules or inhibiting negative regulatory molecules of immune cell activation). In some embodiments, such conjugates are capable of recognizing, binding to, and/or modulating (e.g., inhibiting or activating) immune cells (e.g., NK cells, monocytes/macrophages, dendritic cells, T cells, B cells).

The antibody targeting moieties of the present invention include antibody fragments, including but not limited to Fab, Fab' and F(ab')2, Fd, single chain Fv (scFv), single chain antibodies, disulfide-linked Fv (sdFv), and fragments containing either a VL or VH domain. Antigen-binding antibody fragments, including single chain antibodies, may contain the variable region alone or in combination with all or a portion of the following: hinge region, CH1, CH2, and CH3 domains. Also included in the present invention are antigen-binding fragments that also contain any combination of the variable region and hinge region, CH1, CH2, and CH3 domains. Also included in the present invention are Fc fragments, antigen-Fc fusion proteins, and Fc-targeting moiety conjugates or fusion products (Fc-peptides, Fc-aptamers). The antibody targeting moieties of the present invention may be from any animal origin, including birds and mammals. In one aspect, the antibody targeting moiety is human, murine (e.g., mouse and rat), donkey, sheep, rabbit, goat, guinea pig, camel, horse, or chicken. Additionally, such antibodies may be humanized versions of animal antibodies. The antibody targeting moieties of the present invention may be monospecific, bispecific, trispecific, or of higher order multispecificity.

In various embodiments, the antibody/targeting moiety recruits, binds, and/or activates immune cells (e.g., NK cells, monocytes/macrophages, dendritic cells) via interactions between Fc (in the antibody) and Fc receptors (on immune cells) and via a conjugated variant polypeptide or immunomodulatory protein provided herein. In some embodiments, the antibody/targeting moiety recognizes or binds tumor material via a conjugated variant polypeptide or immunomodulatory protein provided herein, localizing to tumor cells and facilitating modulation of immune cells in the vicinity of the tumor.

Examples of antibodies that can be incorporated into the IgSF conjugate include, but are not limited to, antibodies such as cetuximab (IMC-C225; Erbitux®), trastuzumab (Herceptin®), rituximab (Rituxan®; MabThera®), bevacizumab (Avastin®), alemtuzumab (Campath®; Campath-1H®; Mabcampath®), panitumumab (ABX-EGF; Vectibix®), ranibizumab (Lucentis®), ibritumomab, ibritumomab tiuxetan (Zevalin®), tositumomab, iodine I 131 Tositumomab (BEXXAR®), catumaxomab (Removab®), gemtuzumab, gemtuzumab ozogamicin (Mylotarg®), abatacept (CTLA4-Ig; Orencia®), belatacept (L104EA29YIg; LEA29Y; LEA), ipilimumab (MDX-010; MDX-101), tremelimumab (ticilimumab; CP-675,206), PRS-010, PRS-050, aflibercept (VEGF Trap, AVE005), volociximab (M200), F200, MORAb-009, SS1P (CAT-5001), cixutumumab (IMC-A12), matuzumab (EMD72000), nimotuzumab (h-R3), zalutumumab (HuMax-EGFR), necitumumab IMC-11F8, mAb806/ch806, Sym004, mAb-425, Panorex@(17-1A) (murine monoclonal antibody); Panorex@(17-1A) (chimeric murine monoclonal antibody); IDEC-Y2B8 (murine, anti-CD2O MAb); BEC2 (anti-idiotypic MAb, mimicking the GD epitope) (containing BCG); Oncolym (Lym-1 monoclonal antibody); SMART MI95 Ab, humanized 13'I LYM-I (Oncolym), Ovarex (B43.13, anti-idiotypic mouse MAb); MDX-210 (humanized anti-HER-2 bispecific antibody); 3622W94 Mab binding to EGP40 (17-1A) pancarcinoma antigen on adenocarcinoma; anti-VEGF, Zenapax (SMART anti-Tac (IL-2 receptor); SMART MI95 Ab, humanized Ab, humanized); MDX-210 (humanized anti-HER-2 bispecific antibody); MDX-447 (humanized anti-EGF receptor bispecific antibody); NovoMAb-G2 (pancarcinoma-specific Ab); TNT (chimeric MAb against histone antigen); TNT (chimeric MAb against histone antigen); Gliomab-H (Monoclon s-humanized Ab); GNI-250 Mab; EMD-72000 (chimeric EGF antagonist); Lymphoside (humanized LL2 antibody); and MDX-260 bispecific, targeting GD-2, ANA Ab, SMART IDlO Ab, SMART ABL 364 Ab or ImmuRAIT-CEA. As exemplified by the above list, it is routine to produce antibodies against specific target epitopes.

In some embodiments, the antibody targeting moiety is a full-length antibody or antigen-binding fragment thereof that contains an Fc domain. In some embodiments, the variant polypeptide or immunomodulatory protein is conjugated to the Fc portion of the antibody targeting moiety, e.g., by conjugation to the N-terminus of the Fc portion of the antibody.

In some embodiments, vIgD is linked, directly or indirectly, to the N-terminus or C-terminus of the antibody light and/or heavy chain. In some embodiments, the linkage can be via a peptide linker, such as any of those described above. Various configurations can be constructed. Figures 6A-6C show exemplary configurations. In some embodiments, the antibody conjugate can be produced by co-expression of the antibody heavy and light chains in a cell.

In one aspect of the invention, the targeting agent is an aptamer molecule. For example, in some embodiments, the aptamer is composed of a nucleic acid that functions as the targeting agent. In various embodiments, the IgSF conjugates of the invention include an aptamer specific for a molecule on a tumor cell, tumor vasculature, and/or tumor microenvironment. In some embodiments, the aptamer itself can include a biologically active sequence in addition to the targeting module (sequence), where the biologically active sequence can induce an immune response against the target cell. In other words, such an aptamer molecule is a dual-use agent. In some embodiments, the IgSF conjugates of the invention include the conjugation of an aptamer to an antibody, where the aptamer and the antibody are specific for binding to separate molecules on the tumor cell, tumor vasculature, tumor microenvironment, and/or immune cell.

The term "aptamer" includes DNA, RNA, or peptides selected based on specific binding properties to a particular molecule. For example, aptamers can be selected to bind to specific genes or gene products in tumor cells, tumor vasculature, tumor microenvironment, and/or immune cells as disclosed herein, where selection is made by methods known in the art and familiar to those of skill in the art.

In some aspects of the invention, the targeting agent is a peptide. For example, a variant polypeptide or an immunomodulatory protein provided herein can be conjugated to a peptide capable of binding to a component of a cancer or tumor cell. Thus, such IgSF conjugates of the invention include peptide targeting agents that bind to cellular components of tumor cells, tumor vasculature, and/or components of the tumor microenvironment. In some embodiments, the targeting agent peptide can be an integrin antagonist or agonist. Integrins, including alpha and beta subunits, include numerous types well known to those of skill in the art.

In one embodiment, the targeting agent is Vvβ3. Integrin Vvβ3 is expressed on a wide variety of cells and has been shown to mediate several biologically relevant processes including osteoclast adhesion to bone matrix, vascular smooth muscle cell migration, and angiogenesis. Suitable targeting molecules for integrins include RGD peptides or peptidomimetics as well as non-RGD peptides or peptidomimetics (see, e.g., U.S. Patent Nos. 5,767,071 and 5,780,426) for other integrins such as V4.βi (VLA-4), V4-P7 (see, e.g., U.S. Patent Nos. 6,365,619; Chang et al, Bioorganic & Medicinal Chem Lett, 12:159-163 (2002); Lin et al., Bioorganic & Medicinal Chem Lett, 12:133-136 (2002)).

In some embodiments, an IgSF conjugate is provided that includes a variant polypeptide or an immunomodulatory protein provided herein conjugated to a therapeutic agent. In some embodiments, the therapeutic agent includes, for example, daunomycin, doxorubicin, methotrexate, and vindesine (Rowland et al., Cancer Immunol. Immunother. 21:183-187, 1986). In some embodiments, the therapeutic agent has intracellular activity. In some embodiments, the IgSF conjugate is internalized and the therapeutic agent is a cytotoxin that blocks protein synthesis in the cell, leading to cell death therein. In some embodiments, the therapeutic agent is a cytotoxin that includes a polypeptide having ribosome inactivating activity, including, for example, gelonin, bouganin, saporin, ricin, ricin A chain, bryodin, diphtheria toxin, restrictocin, Pseudomonas exotoxin A and variants thereof. In some embodiments, when the therapeutic agent is a cytotoxin that includes a polypeptide with ribosome-inactivating activity, the IgSF conjugate must be internalized upon binding to the target cell in order for the protein to be cytotoxic to the cell.

In some embodiments, an IgSF conjugate is provided that comprises a variant polypeptide or an immunomodulatory protein provided herein conjugated to a toxin. In some embodiments, the toxin is, for example, a bacterial toxin, such as diphtheria toxin, a plant toxin, such as ricin, a small molecule toxin, such as geldanamycin (Mandler et al., J.Nat. Cancer Inst. 92 (19):1573-1581 (2000); Mandler et al., Bioorganic & Med. Chem. Letters 10:1025-1028 (2000); Mandler et al., Bioconjugate Chem. 13:786-791 (2002)), maytansinoids (EP 1391213; Liu et al., Proc. Natl. Acad. Sci. USA 93:8618-8623 (1996)), and calicheamicin (Lode et al., Cancer Res. 58:2928 (1998); Hinman et al., Cancer Res. 53:3336-3342 (1993). The toxins may exert their cytotoxic and cytostatic effects by mechanisms including tubulin binding, DNA binding, or topoisomerase inhibition.

In some embodiments, IgSF conjugates are provided that include a variant polypeptide or an immunomodulatory protein provided herein conjugated to a label capable of indirectly or directly generating a detectable signal. These IgSF conjugates can be used for research or diagnostic applications, such as for in vivo detection of cancer. The label is preferably capable of producing a detectable signal, either directly or indirectly. For example, the label can be a radiopaque or radioactive isotope, such as 3H, 14C, 32P, 35S, 123I, 125I, 131I; a fluorescent (fluorophore) or chemiluminescent (chromophore) compound, such as fluorescein isothiocyanate, rhodamine, or luciferin; an enzyme, such as alkaline phosphatase, β-galactosidase, or horseradish peroxidase; an imaging agent; or a metal ion. In some embodiments, the label is a radioactive atom, such as 99Tc or 123I, for scintigraphy studies, or a spin label, such as zirconium-89, iodine-123, iodine-131, indium-111, fluorine-19, carbon-13, nitrogen-15, oxygen-17, gadolinium, manganese or iron, for nuclear magnetic resonance (NMR) imaging (also known as magnetic resonance imaging, MRI). Zirconium-89 may be complexed to various metal chelators and conjugated to antibodies, for example, for PET imaging (WO 2011/056983). In some embodiments, the IgSF conjugate is indirectly detectable. For example, the IgSF conjugate can be detected using a secondary antibody specific for the IgSF conjugate and containing a detectable label.

IgSF conjugates may be prepared using any method known in the art. See, e.g., WO 2009/067800, WO 2011/133886, and U.S. Patent Application Publication No. 2014322129, which are incorporated by reference in their entireties.

The variant polypeptide or immunomodulatory protein of the IgSF conjugate may be "linked" to the effector moiety by any means by which the variant polypeptide or immunomodulatory protein can associate with or be linked to the effector moiety. For example, the variant polypeptide or immunomodulatory protein of the IgSF conjugate may be linked to the effector moiety by chemical or recombinant means. Chemical means for preparing fusions or conjugates are known in the art and can be used to prepare the IgSF conjugates. The method used to conjugate the variant polypeptide or immunomodulatory protein with the effector moiety must be capable of connecting the variant polypeptide or immunomodulatory protein with the effector moiety without interfering with the ability of the variant polypeptide or immunomodulatory protein to bind to their one or more counter structure ligands.

The variant polypeptide or immunomodulatory protein of the IgSF conjugate may be indirectly linked to the effector moiety. For example, the variant polypeptide or immunomodulatory protein of the IgSF conjugate may be directly linked to a liposome containing one of several effector moieties. The effector moiety and/or the variant polypeptide or immunomodulatory protein may also be bound to a solid surface.

In some embodiments, the variant polypeptide or immunomodulatory protein and the effector moiety of the IgSF conjugate are both proteins and can be conjugated using techniques well known in the art. There are hundreds of crosslinkers available that can conjugate two proteins (see, for example, "Chemistry of Protein Conjugation and Crosslinking," 1991, Shans Wong, CRC Press, Ann Arbor). Crosslinkers are generally selected based on reactive functional groups available on or inserted into the variant polypeptide or immunomodulatory protein and/or effector moiety. In addition, if there are no reactive groups, photoactivatable crosslinkers can be used. In certain cases, it may be desirable to include a spacer between the variant polypeptide or immunomodulatory protein and the effector moiety. Cross-linking agents known in the art include the homobifunctional agents: glutaraldehyde, dimethyladipimidate, and bis(diazobenzidine), and the heterobifunctional agents: m-maleimidobenzoyl-N-hydroxysuccinimide and sulfo-m-maleimidobenzoyl-N-hydroxysuccinimide.

In some embodiments, the variant polypeptide or immunomodulatory protein of the IgSF conjugate may be modified at specific residues for chemical attachment of the effector moiety. Specific residues used for chemical attachment of molecules known in the art include lysine and cysteine. Crosslinkers are selected based on the reactive functional groups inserted on the variant polypeptide or immunomodulatory protein and the reactive functional groups available on the effector moiety.

IgSF conjugates may also be prepared using recombinant DNA techniques. In such cases, a DNA sequence encoding a variant polypeptide or an immunomodulatory protein is fused to a DNA sequence encoding an effector moiety, resulting in a chimeric DNA molecule. The chimeric DNA sequence is transfected into a host cell that expresses the fusion protein. The fusion protein can be recovered from the cell culture and purified using techniques known in the art.

Examples of conjugating a label effector moiety to a variant polypeptide or immunomodulatory protein include those described in Hunter, et al., Nature 144:945 (1962); David, et al., Biochemistry 13:1014 (1974); Pain, et al., J. Immunol. Meth. 40:219 (1981); Nygren, J. Histochem. and Cytochem. 30:407 (1982); Wensel and Meares, Radioimmunoimaging And Radioimmunotherapy, Elsevier, N.Y. (1983); and Colcher et al., “Use Of Monoclonal Antibodies As Radiopharmaceuticals For The Localization Of Human Carcinoma Xenografts In Athymic Mice”, Meth. Enzymol., 121:802-16 (1986).

Radiolabels or other labels may be incorporated into the conjugates by known methods. For example, peptides may be biosynthesized or synthesized by chemical amino acid synthesis using suitable amino acid precursors, e.g., containing fluorine-19 instead of hydrogen. Labels such as 99Tc or 123I, 186Re, 188Re and 111In may be attached via cysteine residues in the peptide. Yttrium-90 may be attached via lysine residues. Iodine-123 may be incorporated using the IODOGEN method (Fraker et al., Biochem. Biophys. Res. Commun. 80:49-57 (1978)). "Monoclonal Antibodies in Immunoscintigraphy" (Chatal, CRC Press 1989) describes other methods in detail.

Conjugates of variant polypeptides or immunomodulatory proteins and cytotoxic agents may be prepared using a wide variety of bifunctional protein coupling agents, such as N-succinimidyl-3-(2-pyridyldithio)propionate (SPDP), succinimidyl-4-(N-maleimidomethyl)cyclohexane-1-carboxylate (SMCC), iminothiolane (IT), bifunctional derivatives of imidoesters (e.g., dimethyl adipimidate HCI), active esters (e.g., disuccinimidyl suberate), aldehydes (e.g., glutaraldehyde), bis-azido compounds (e.g., bis(p-azidobenzoyl)hexanediamine), bis-diazonium derivatives (e.g., bis-(p-diazoniumbenzoyl)-ethylenediamine), diisocyanates (e.g., toluene 2,6-diisocyanate), and bis-active fluorine compounds (e.g., 1,5-difluoro-2,4-dinitrobenzene). For example, ricin immunotoxins can be prepared as described in Vitetta et al., Science 238:1098 (1987). Carbon-14 labeled 1-isothiocyanatobenzyl-3-methyldiethylenetriaminepentaacetic acid (MX-DTPA) is an exemplary chelating agent for conjugation of radionucleotides to antibodies. See, for example, WO 94/11026. The linker can be a "cleavable linker" that facilitates release of the cytotoxic drug in the cell. For example, acid-labile linkers, peptidase-sensitive linkers, photolabile linkers, dimethyl linkers or disulfide-containing linkers (Chari et al., Cancer Research 52:127-131 (1992); U.S. Patent No. 5,208,020) may be used.

The IgSF conjugates of the present invention expressly contemplate, but are not limited to, drug conjugates prepared with cross-linker reagents: BMPS, EMCS, GMBS, HBVS, LC-SMCC, MBS, MPBH, SBAP, SIA, SIAB, SMCC, SMPB, SMPH, sulfo-EMCS, sulfo-GMBS, sulfo-KMUS, sulfo-MBS, sulfo-SIAB, sulfo-SMCC, and sulfo-SMPB, and SVSB (succinimidyl-(4-vinylsulfone)benzoate), available commercially (e.g., from Pierce Biotechnology, Inc., Rockford, IL, U.S.A.). See pages 467-498 of Applications Handbook and Catalog 2003-2004.

D. Transmembrane and Secreted Immunomodulatory Proteins and Modified Cells Modified cells (alternatively, "modified cells") expressing immunomodulatory variant CD80 polypeptides are provided herein. In some embodiments, the expressed immunomodulatory variant CD80 polypeptide is a transmembrane protein and is expressed on the surface. In some embodiments, the expressed immunomodulatory variant CD80 polypeptide is expressed in and secreted from the cell.

1. Transmembrane immunomodulatory proteins In some embodiments, the immunomodulatory polypeptide comprising variant CD80 can be a membrane-bound protein. As described in more detail below, the immunomodulatory polypeptide can be a transmembrane immunomodulatory polypeptide comprising variant CD80, containing an ectodomain containing at least one affinity-modified IgSF domain (IgV or IgC), a transmembrane domain, and optionally a cytoplasmic domain. In some embodiments, the transmembrane immunomodulatory protein can be expressed on the surface of an immune cell, such as a mammalian cell, including the surface of a lymphocyte (e.g., a T cell or a NK cell) or an antigen-presenting cell. In some embodiments, the transmembrane immunomodulatory protein is expressed on the surface of a mammalian T cell, including a T cell such as a helper T cell, a cytotoxic T cell (or a cytotoxic T lymphocyte or CTL), a natural killer T cell, a regulatory T cell, a memory T cell, or a gamma delta T cell. In some embodiments, the mammalian cell is an antigen-presenting cell (APC). Typically, but not exclusively, the ectodomain (or alternatively, the "extracellular domain") comprises one or more amino acid modifications (e.g., amino acid substitutions) of a variant CD80 of the invention. Thus, for example, in some embodiments, a transmembrane protein comprises an ectodomain that comprises one or more amino acid substitutions of a variant CD80 of the invention.

In some embodiments, the modified cells express a variant CD80 polypeptide that is a transmembrane immunomodulatory polypeptide (TIP), which can be a membrane protein, such as a transmembrane protein. In typical embodiments, the ectodomain of the membrane protein comprises the extracellular domain of a variant CD80 provided herein or an IgSF domain thereof, which contains one or more amino acid substitutions in at least one IgSF domain as described. The transmembrane immunomodulatory protein provided herein further contains a transmembrane domain linked to the ectodomain. In some embodiments, the transmembrane domain provides the encoded protein for cell surface expression on the cell. In some embodiments, the transmembrane domain is directly linked to the ectodomain. In some embodiments, the transmembrane domain is indirectly linked to the ectodomain via one or more linkers or spacers. In some embodiments, the transmembrane domain contains predominantly hydrophobic amino acid residues, such as leucine and valine.

In some embodiments, a full-length transmembrane anchor domain can be used to ensure that the TIP is expressed on the surface of the engineered cell, e.g., engineered T cell. Conveniently, this can be derived from the particular native protein for which affinity is being engineered (e.g., CD80 or other native IgSF protein) and can simply be fused to the sequence of the first membrane proximal domain in the same manner as a native IgSF protein (e.g., CD80). In some embodiments, the transmembrane immunomodulatory protein comprises the transmembrane domain of the corresponding wild-type or unmodified IgSF member (e.g., the transmembrane domain contained in the amino acid sequence shown in SEQ ID NO:1) (Table 2). In some embodiments, the membrane-bound form comprises the transmembrane domain of the corresponding wild-type or unmodified polypeptide, e.g., corresponding to residues 243-263 of SEQ ID NO:1.

In some embodiments, the transmembrane domain is a non-native transmembrane domain that is not the transmembrane domain of native CD80. In some embodiments, the transmembrane domain is derived from a transmembrane domain from another non-CD80 family member polypeptide that is a membrane-bound or transmembrane protein. In some embodiments, a transmembrane anchor domain from another protein on a T cell can be used. In some embodiments, the transmembrane domain is derived from CD8. In some embodiments, the transmembrane domain can further contain an extracellular portion of CD8 that serves as a spacer domain. Exemplary CD8-derived transmembrane domains are set forth in SEQ ID NO:246 or 399 or portions thereof that contain the CD8 transmembrane domain. In some embodiments, the transmembrane domain is a synthetic transmembrane domain.

In some embodiments, the transmembrane immunomodulatory protein further contains an endodomain, e.g., a cytoplasmic signaling domain, linked to the transmembrane domain. In some embodiments, the cytoplasmic signaling domain induces cell signaling. In some embodiments, the endodomain of the transmembrane immunomodulatory protein comprises the cytoplasmic domain of a corresponding wild-type or unmodified polypeptide, e.g., the cytoplasmic domain contained in the amino acid sequence shown in SEQ ID NO:5 (see Table 2).

In some embodiments, the provided transmembrane immunomodulatory proteins are or include variant CD80s and include an amino acid sequence that exhibits at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% sequence identity to SEQ ID NO:253 and contain an ectodomain and a transmembrane domain that includes at least one affinity modified CD80 IgSF domain as described. In some embodiments, the transmembrane immunomodulatory proteins contain any one or more amino acid substitutions in the IgSF domain (e.g., IgV domain) as described, including any shown in Table 1. In some embodiments, the transmembrane immunomodulatory proteins can further include a cytoplasmic domain as described. In some embodiments, the transmembrane immunomodulatory proteins can further include a signal peptide. In some embodiments, the signal peptide is the native signal peptide of a wild-type IgSF member, e.g., contained in the amino acid sequence set forth in SEQ ID NO:1 (see, e.g., Table 2).

Also provided are nucleic acid molecules encoding such transmembrane immunomodulatory proteins. In some embodiments, the nucleic acid molecule encoding the transmembrane immunomodulatory protein comprises a nucleotide sequence encoding an amino acid sequence exhibiting at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% sequence identity to SEQ ID NO:253 and contains an ectodomain comprising at least one affinity-modified IgSF domain as described, a transmembrane domain, and optionally a cytoplasmic domain. In some embodiments, the nucleic acid molecule can further comprise a sequence of nucleotides encoding a signal peptide. In some embodiments, the signal peptide is the native signal peptide of the corresponding wild-type IgSF member (see, e.g., Table 2).

An exemplary transmembrane immunomodulatory protein is CD80 TIP, comprising i) the amino acid sequence set forth in SEQ ID NO:241, or ii) an amino acid sequence that exhibits at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% sequence identity to SEQ ID NO:241 and includes an affinity-modified domain of SEQ ID NO:241 or an amino acid substitution contained therein. Also provided is i) a nucleotide sequence as set forth in SEQ ID NO:242, ii) a sequence that exhibits at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% sequence identity to SEQ ID NO:242 and encodes a TIP containing an affinity-modified domain of SEQ ID NO:241 or a polypeptide containing amino acid substitutions of SEQ ID NO:241, or iii) a sequence of i) or ii) having degenerate codons.

In some embodiments, a CAR-associated transmembrane immunomodulatory protein is provided, wherein the endodomain of the transmembrane immunomodulatory protein comprises a cytoplasmic signaling domain that comprises at least one ITAM (immunoreceptor tyrosine-based activation motif)-containing signaling domain. ITAM is a conserved motif found in many protein signaling domains involved in immune cell signaling, including the CD3-zeta chain ("CD3-z"), which is involved in T cell receptor signaling. In some embodiments, the endodomain comprises a CD3-ζ signaling domain. In some embodiments, the CD3-ζ signaling domain comprises an amino acid sequence as set forth in SEQ ID NO:333, or an amino acid sequence that exhibits at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% of SEQ ID NO:247 and retains activity in T cell signaling. In some embodiments, the endodomain of the CAR-associated transmembrane immunomodulatory protein can further comprise a costimulatory signaling domain that further modulates the immunomodulatory response of T cells. In some embodiments, the costimulatory signaling domain is CD28, ICOS, 41BB or OX40. In some embodiments, the costimulatory signaling domain is derived from CD28 or 4-1BB and comprises an amino acid sequence set forth in any of SEQ ID NOs:400-403, or an amino acid sequence that exhibits at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% of SEQ ID NOs:400-403 and retains activity of T cell costimulatory signaling. In some embodiments, the CAR-associated transmembrane immunomodulatory protein provided has the characteristic of a CAR, stimulating T cell signaling by binding of the affinity engineered IgSF domain to a cognate binding partner or counter structure. In some embodiments, specific binding of an affinity engineered IgSF domain to its counter structure can lead to altered immune activity of T cell activity, as reflected by changes in cytotoxicity, proliferation or cytokine production.

In some embodiments, the transmembrane immunomodulatory protein does not contain an endodomain capable of mediating cytoplasmic signaling. In some embodiments, the transmembrane immunomodulatory protein lacks the signaling mechanism of the wild-type or unmodified polypeptide and therefore does not induce cell signaling by itself. In some embodiments, the transmembrane immunomodulatory protein lacks the intracellular (cytoplasmic) domain or a portion of the intracellular domain of the corresponding wild-type or unmodified polypeptide, e.g., the cytoplasmic signaling domain contained in the amino acid sequence shown in SEQ ID NO:5 (see Table 2). In some embodiments, the transmembrane immunomodulatory protein does not contain an ITIM (immunoreceptor tyrosine-based inhibitory motif), as contained in certain inhibitory receptors, including, for example, inhibitory receptors of the IgSF family (e.g., PD-1 or TIGIT). Thus, in some embodiments, the transmembrane immunomodulatory protein contains only an ectodomain and a transmembrane domain (e.g., any as described).

2. Secreted immunomodulatory proteins and modified cells In some embodiments, the CD80 variant immunomodulatory polypeptides containing any one or more of the amino acid mutations as described herein are secretable, for example, when expressed in a cell. Such variant CD80 immunomodulatory proteins do not contain a transmembrane domain. In some embodiments, the variant CD80 immunomodulatory proteins are not conjugated to a half-life extending moiety (e.g., an Fc domain or a multimerization domain). In some embodiments, the variant CD80 immunomodulatory proteins contain a signal peptide, such as an antibody signal peptide or other efficient signal sequence, that directs the domain outside the cell. When the immunomodulatory protein contains a signal peptide and is expressed by a modified cell, the signal peptide allows the immunomodulatory protein to be secreted by the modified cell. Generally, the signal peptide, or a portion of the signal peptide, is cleaved from the immunomodulatory protein upon secretion. The immunomodulatory protein can be encoded by a nucleic acid, which can be part of an expression vector. In some embodiments, the immunomodulatory protein is expressed and secreted by a cell (e.g., an immune cell, such as a primary immune cell).

Thus, in some embodiments, variant CD80 immunomodulatory proteins are provided that further comprise a signal peptide. In some embodiments, provided herein are nucleic acid molecules that encode variant CD80 immunomodulatory proteins operably linked to a secretory sequence encoding a signal peptide.

A signal peptide is a sequence on the N-terminus of an immunomodulatory protein that signals secretion of the immunomodulatory protein from a cell. In some embodiments, the amino acid length of the signal peptide is about 5 to about 40 amino acids (e.g., about 5 to about 7, about 7 to about 10, about 10 to about 15, about 15 to about 20, about 20 to about 25, or about 25 to about 30, about 30 to about 35, or about 35 to about 40 amino acids).

In some embodiments, the signal peptide is a native signal peptide from the corresponding wild-type CD80 (see Table 2). In some embodiments, the signal peptide is a non-native signal peptide. For example, in some embodiments, the non-native signal peptide is a variant native signal peptide from the corresponding wild-type CD80 and can include one or more (e.g., 2, 3, 4, 5, 6, 7, 8, 9, or 10 or more) substitutions, insertions, or deletions. In some embodiments, the non-native signal peptide is a signal peptide of a family member from the same IgSF family as the wild-type IgSF family member, or a variant thereof. In some embodiments, the non-native signal peptide is a signal peptide of an IgSF family member from an IgSF family different from the wild-type IgSF family member, or a variant thereof. In some embodiments, the signal peptide is a signal peptide from a non-IgSF protein family or a variant thereof, such as a signal peptide from an immunoglobulin (e.g., IgG heavy chain or IgGκ light chain), a cytokine (e.g., interleukin-2 (IL-2), or CD33), a serum albumin protein (e.g., HSA or albumin), a human azurocidin preprotein signal sequence, luciferase, trypsinogen (e.g., chymotrypsinogen or trypsinogen), or other signal peptide capable of efficiently secreting a protein from a cell. Exemplary signal peptides include any of those listed in Table 6.

Table 6. Exemplary signal peptides

In some embodiments of a secreted variant CD80 immunomodulatory protein, the immunomodulatory protein includes a signal peptide when expressed, and the signal peptide (or a portion thereof) is cleaved from the immunomodulatory protein upon secretion.

In some embodiments, the modified cells express a variant CD80 polypeptide that is secreted from the cell. In some embodiments, such variant CD80 polypeptides are encoded by a nucleic acid molecule encoding an immunomodulatory protein under operative control of a signal sequence for secretion. In some embodiments, the encoded immunomodulatory protein is secreted when expressed from the cell. In some embodiments, the immunomodulatory protein encoded by the nucleic acid molecule does not include a transmembrane domain. In some embodiments, the immunomodulatory protein encoded by the nucleic acid molecule does not include a half-life extending moiety (e.g., an Fc domain or a multimerization domain). In some embodiments, the immunomodulatory protein encoded by the nucleic acid molecule includes a signal peptide. In some embodiments, the nucleic acid of the invention further includes a nucleotide sequence encoding a secretory or signal peptide operably linked to the nucleic acid encoding the immunomodulatory protein, thereby allowing for secretion of the immunomodulatory protein.

3. Cells and Cellular Modifications Provided herein are modified cells that express any of the provided immunomodulatory polypeptides. In some embodiments, the modified cells express any of the provided transmembrane immunomodulatory polypeptides on their surface. In some embodiments, the modified cells express and are capable of secreting or secreting immunomodulatory proteins from the cell under conditions suitable for the secretion of the protein. In some embodiments, the immunomodulatory proteins are expressed on lymphocytes, such as tumor infiltrating lymphocytes (TILs), T cells or NK cells, or on myeloid cells. In some embodiments, the modified cells are antigen presenting cells (APCs). In some embodiments, the modified cells are modified mammalian T cells or modified mammalian antigen presenting cells (APCs). In some embodiments, the modified T cells or APCs are human or murine cells.

In some embodiments, the modified T cells include, but are not limited to, helper T cells, cytotoxic T cells (alternatively, cytotoxic T lymphocytes or CTLs), natural killer T cells, regulatory T cells, memory T cells, or gamma delta T cells. In some embodiments, the modified T cells are CD4 ⁺ or CD8 ⁺ . In addition to MHC signals, the modified T cells also require costimulatory signals, which in some embodiments are provided by a variant CD80 transmembrane immunomodulatory polypeptide expressed in a membrane-bound form as discussed above.

In some embodiments, modified APCs include, for example, APCs expressing MHC II, such as macrophages, B cells, and dendritic cells, as well as aAPCs, including both cellular and acellular (e.g., biodegradable polymeric microparticles). Artificial APCs (aAPCs) are synthetic versions of APCs that can act in a similar manner to APCs in that they present antigens to T cells and activate them. Antigen presentation is carried out by MHC (class I or class II). In some embodiments, in modified APCs such as aAPCs, the antigen loaded onto the MHC is, in some embodiments, a tumor-specific or tumor-associated antigen. The antigen loaded onto the MHC is recognized by the T cell receptor (TCR) of a T cell, which in some cases can express CTLA-4 or CD80 or other molecules recognized by the variant CD80 polypeptides provided herein. Materials that can be used to modify aAPCs include poly(glycolic acid), poly(lactic-co-glycolic acid), iron oxide, liposomes, lipid bilayers, sepharose, and polystyrene.

In some embodiments, the immunomodulatory proteins provided herein, e.g., transmembrane or secreted immunomodulatory proteins, are co-expressed or modified in cells expressing an antigen-binding receptor, e.g., a recombinant receptor, e.g., a chimeric antigen receptor (CAR) or a T cell receptor (TCR). In some embodiments, the modified cells, e.g., modified T cells, recognize a desired antigen associated with cancer, inflammatory and autoimmune disorders, or viral infection. In a specific embodiment, the antigen-binding receptor contains an antigen-binding moiety that specifically binds to a tumor-specific or tumor-associated antigen. In some embodiments, the modified T cells are CAR (chimeric antigen receptor) T cells that contain an antigen-binding domain (e.g., scFv) that specifically binds to an antigen, e.g., a tumor-specific or tumor-associated antigen. In some embodiments, the antigen-binding domain (e.g., scFv) is specific for CD19. An exemplary CAR is an anti-CD19 CAR, e.g., a CAR containing an anti-CD19 scFv as set forth in SEQ ID NO:245. In some embodiments, a TIP protein is expressed in modified T cell receptor cells or and modified chimeric antigen receptor cells. In such embodiments, the modified cells co-express TIP and CAR or TCR.

In some embodiments, the CAR further contains an intracellular signaling domain (endodomain) including a spacer or hinge, a transmembrane domain, and an ITAM signaling domain (e.g., a CD3ζ signaling domain). In some embodiments, the CAR further comprises a costimulatory signaling domain. In some embodiments, the spacer or hinge is between the antigen binding domain and the transmembrane domain, e.g., between the antigen binding domain and the plasma membrane when expressed on a cell. In some embodiments, the spacer or hinge is derived from an IgG subclass (e.g., IgG1 and IgG4), IgD, or CD8 (see, e.g., Qin et al. (2017) J. Hematol. Oncol., 10:68). In some embodiments, the spacer or hinge is derived from IgG1.

In some embodiments, the spacer and transmembrane domains are hinge and transmembrane domains derived from CD8, e.g., as set forth in SEQ ID NO:246 or 399, or an amino acid sequence exhibiting at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or higher sequence identity to SEQ ID NO:332 or 364. In some embodiments, the endodomain comprises a CD3-zeta signaling domain. In some embodiments, the CD3-zeta signaling domain comprises an amino acid sequence as set forth in SEQ ID NO:247, or an amino acid sequence that exhibits at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% of SEQ ID NO:247 and retains T cell signaling activity. In some embodiments, the endodomain of the CAR-associated transmembrane immunomodulatory protein can further comprise a costimulatory signaling domain that further modulates the immunomodulatory response of T cells. In some embodiments, the costimulatory signaling domain is CD28, ICOS, 41BB or OX40. In some embodiments, the costimulatory signaling domain is derived from CD28 or 4-1BB and comprises an amino acid sequence set forth in any of SEQ ID NOs: 400-403, or an amino acid sequence that exhibits at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% of SEQ ID NOs: 400-403 and retains T cell costimulatory signaling activity.

In another embodiment, the modified T cell possesses a TCR, including a recombinant or modified TCR. In some embodiments, the TCR can be a native TCR. One of skill in the art will recognize that native mammalian T cell receptors generally contain alpha and beta chains (or gamma and delta chains) that are involved in antigen-specific recognition and binding. In some embodiments, the TCR is a modified TCR that has been modified. In some embodiments, the TCR of the modified T cell specifically binds to a tumor-associated or tumor-specific antigen presented by an APC.

In some embodiments, the immunomodulatory polypeptide, e.g., a transmembrane immunomodulatory polypeptide or a secreted immunomodulatory polypeptide, can be incorporated into the modified cell, e.g., a modified T cell or modified APC, by a wide variety of strategies, such as those employed in recombinant host cells. A wide variety of methods are known in the art for introducing DNA constructs into primary T cells. In some embodiments, viral transduction or plasmid electroporation is employed. In typical embodiments, the nucleic acid molecule encoding the immunomodulatory protein, or the expression vector, contains a signal peptide for localizing the expressed transmembrane immunomodulatory protein to the cell membrane or for secretion. In some embodiments, the nucleic acid encoding the transmembrane immunomodulatory protein of the invention is subcloned into a viral vector, e.g., a retroviral vector, that allows expression in a host mammalian cell. The expression vector can be introduced into a mammalian host cell, and under host cell culture conditions, the immunomodulatory protein is expressed on the surface or secreted.

In an illustrative example, primary T cells can be purified ex vivo (CD4 or CD8 cells or both) and stimulated with an activation protocol consisting of beads coated with various TCR/CD28 agonists, e.g., anti-CD3/anti-CD28. After a 2 or 3 day activation process, recombinant expression vectors containing immune modulating polypeptides can be stably introduced into the primary T cells by lentiviral or retroviral transduction protocols or plasmid electroporation strategies standard in the art. Cells can be monitored for immune modulating polypeptide expression, for example, by flow cytometry using anti-epitope tags or antibodies that cross-react with the native parent molecule and polypeptide (including variant CD80). T cells expressing immune modulating polypeptides can be enriched by sorting with anti-epitope tag antibodies or enriched for high or low expression, depending on the application.

The immune-modulating polypeptide expression allows the modified T cells to be assayed for appropriate function by a wide variety of means. Co-expression of the modified CAR or TCR can be verified to show that this portion of the modified T cells was largely unaffected by the expression of the immune-modulating protein. Once verified, standard in vitro cytotoxicity, proliferation, or cytokine assays (e.g., IFN-γ expression) can be used to assess the functionality of the modified T cells. Exemplary standard endpoints are the rate of lysis of the tumor line, the proliferation of the modified T cells, or IFN-γ protein expression in the culture supernatant. Modified constructs that result in a statistically significant increase in lysis of the tumor line, an increase in proliferation of the modified T cells, or an increase in IFN-γ expression relative to the control construct can be selected. Additionally, unmodified cells, e.g., natural primary or endogenous T cells, can also be incorporated into the same in vitro assay to measure the ability of the immune-modulating polypeptide constructs expressed on the modified cells, e.g., the modified T cells, to modulate activity in bystander natural T cells, including, in some cases, activating and generating effector functions. Increased expression of activation markers, such as CD69, CD44, or CD62L, can also be monitored on endogenous T cells, and increased proliferation and/or cytokine production can indicate the desired activity of the immunomodulatory protein expressed on the modified T cells.

In some embodiments, similar assays can be used to compare the function of modified T cells containing CAR or TCR alone with those containing CAR or TCR and TIP constructs. Typically, these in vitro assays are performed by plating "tumor" cell lines containing various ratios of modified T cells and cognate CAR or TCR antigens together in culture. Standard endpoints are the lysis rate of tumor lines, the proliferation of modified T cells, or IFN-γ production in the culture supernatant. Modified immunomodulatory proteins that resulted in a statistically significant increase in tumor line lysis, increased proliferation of modified T cells, or increased IFN-γ production relative to the same TCR or CAR construct alone can be selected. Modified human T cells can be analyzed in immunodeficient mice, such as the NSG strain, which lacks mouse T, NK, and B cells. Modified human T cells, in which CAR or TCR binds to a target counter structure on the xenograft and co-expresses the affinity-modified IgSF domain of TIP, can be adoptively transferred in vivo at different cell numbers and ratios compared to the xenograft. For example, the engraftment of CD19 ⁺ leukemia tumor lines containing luciferase/GFP vectors can be monitored by bioluminescence or ex vivo by flow cytometry.In one common embodiment, xenografts are introduced into murine models, followed by modified T cells several days later.Modified T cells containing immunomodulatory proteins can be assayed for increased survival, tumor clearance, or proliferation numbers of modified T cells compared to modified T cells containing CAR or TCR alone.Similar to in vitro assays, endogenous natural (i.e., unmodified) human T cells can also be co-adoptively transferred to predict the success of epitope spreading in the population, which leads to better survival or tumor clearance.

E. Infectious agents expressing variant polypeptides and immunomodulatory proteins Also provided are infectious agents containing nucleic acids encoding any of the variant polypeptides, including secretory or transmembrane immunomodulatory proteins described herein, such as CD80 vIgD polypeptides. In some embodiments, such infectious agents can deliver nucleic acids encoding variant immunomodulatory polypeptides, such as CD80 vIgD polypeptides, described herein, to target cells of a subject, such as immune cells and/or antigen-presenting cells (APCs) or tumor cells of a subject. Also provided are nucleic acids contained in such infectious agents and/or nucleic acids, such as vectors and/or plasmids, for the generation or modification of such infectious agents, as well as compositions containing such infectious agents.

In some embodiments, the infectious agent is a microorganism or pathogen. In some embodiments, the infectious agent is a virus or bacteria. In some embodiments, the infectious agent is a virus. In some embodiments, the infectious agent is a bacteria. In some embodiments, such an infectious agent can deliver a nucleic acid sequence encoding any of the variant polypeptides, including secreted or transmembrane immunomodulatory proteins described herein, such as CD80 vIgD polypeptides. Thus, in some embodiments, cells of a subject infected or contacted with an infectious agent can express or secrete a variant immunomodulatory polypeptide on the cell surface. In some embodiments, the infectious agent can also deliver a nucleic acid encoding one or more other therapeutic agents or other therapeutic agents to cells and/or the environment within the subject. In some embodiments, other therapeutic agents that can be delivered by the infectious agent include cytokines or other immunomodulatory molecules.

In some embodiments, the infectious agent, e.g., a virus or bacterium, contains a nucleic acid sequence encoding any of the variant polypeptides comprising a secreted or transmembrane immunomodulatory protein described herein, e.g., a CD80 vIgD polypeptide, and upon contact and/or infection of a cell of the subject, the cell expresses the variant polypeptide comprising a secreted or transmembrane immunomodulatory protein, e.g., a CD80 vIgD polypeptide, encoded by the nucleic acid sequence contained in the infectious agent. In some embodiments, the infectious agent can be administered to the subject. In some embodiments, the infectious agent can be transduced ex vivo into cells from the subject.

In some embodiments, the variant polypeptides comprising transmembrane immunomodulatory proteins expressed by cells infected with an infectious agent, e.g., CD80 vIgD polypeptides, are transmembrane proteins and expressed on the surface. In some embodiments, the variant polypeptides comprising secreted immunomodulatory proteins expressed by cells infected with an infectious agent, e.g., CD80 vIgD polypeptides, are expressed and secreted from the cells. The transmembrane immunomodulatory proteins or secreted immunomodulatory proteins can be any of those described herein.

In some embodiments, the cells of interest targeted by the infectious agent include tumor cells, immune cells, and/or antigen-presenting cells (APCs). In some embodiments, the infectious agent targets cells in the tumor microenvironment (TME). In some embodiments, the infectious agent delivers a nucleic acid encoding a variant polypeptide, including a secreted or transmembrane immunomodulatory protein, such as a CD80 vIgD polypeptide, to an appropriate cell (e.g., an APC, e.g., a cell that presents a peptide/MHC complex on its cell surface, e.g., a dendritic cell) or tissue (e.g., a lymphoid tissue), thereby inducing and/or enhancing a desired effect, e.g., immunomodulatory and/or specific cell-mediated immune response, e.g., CD4 and/or CD8 T cell response, which may include a cytotoxic T cell (CTL) response. In some embodiments, the infectious agent targets an APC, e.g., a dendritic cell (DC). In some embodiments, the nucleic acid molecule delivered by the infectious agent described herein includes appropriate nucleic acid sequences, e.g., regulatory elements, e.g., promoters, necessary for expression in a particular target cell of an operably linked coding sequence encoding a variant immunomodulatory polypeptide.

In some embodiments, the infectious agent containing a nucleic acid sequence encoding an immunomodulatory polypeptide can also contain a nucleic acid sequence encoding one or more additional gene products, e.g., a cytokine, a prodrug converting enzyme, a cytotoxin, and/or a detectable gene product. For example, in some embodiments, the infectious agent is an oncolytic virus, and the virus can include a nucleic acid sequence encoding an additional therapeutic gene product (e.g., Kirn et al., (2009) Nat Rev Cancer 9:64-71; Garcia-Aragoncillo et al., (2010) Curr Opin Mol Ther 12:403-411; U.S. Patent Nos. 7,588,767, 7,588,771, 7,662,398, and 7,754,221, and U.S. Patent Application Publication Nos. 2007/0202572, 2007/0212727, 2010/0062016, 2009/0098529, 2009/0053244, See, e.g., 2009/0155287, 2009/0117034, 2010/0233078, 2009/0162288, 2010/0196325, 2009/0136917, and 2011/0064650. In some embodiments, the additional gene product can result in the death of a target cell (e.g., a tumor cell). The gene product may be a therapeutic gene product capable of enhancing or strengthening or modulating an immune response (e.g., a cytokine). Exemplary gene products are also included among anti-cancer agents, anti-metastatic agents, anti-angiogenic agents, immunomodulatory molecules, immune checkpoint inhibitors, antibodies, cytokines, growth factors, antigens, cytotoxic gene products, pro-apoptotic gene products, anti-apoptotic gene products, cell matrix degrading genes, genes for tissue regeneration, and genes for reprogramming human somatic cells to pluripotency, as well as other genes described herein or known to those of skill in the art. In some embodiments, the additional gene product is granulocyte-macrophage colony stimulating factor (GM-CSF).

1. Virus In some embodiments, the infectious agent is a virus. In some embodiments, the infectious agent is an oncolytic virus or a virus that targets specific cells, such as immune cells. In some embodiments, the infectious agent targets tumor cells and/or cancer cells of a subject. In some embodiments, the infectious agent targets immune cells or antigen-presenting cells (APCs).

In some embodiments, the infectious agent is an oncolytic virus. Oncolytic viruses are viruses that accumulate and replicate in tumor cells. By replication in the cells and optional delivery of nucleic acids encoding variant immunomodulatory variant CD80 polypeptides or immunomodulatory proteins described herein, the tumor cells can be lysed and the tumor can shrink and be eliminated. Oncolytic viruses can have a broad host and cell type range. For example, oncolytic viruses accumulate in immune privileged cells or immune privileged tissues, including tumors and/or metastases, and also wound tissues and cells, thus allowing delivery and expression of nucleic acids encoding variant immunomodulatory polypeptides described herein in a broad range of cell types. Oncolytic viruses can also replicate specifically in tumor cells, resulting in tumor cell lysis and efficient tumor regression.

Exemplary oncolytic viruses include adenoviruses, adeno-associated viruses, herpes viruses, herpes simplex viruses, vesicular oral viruses, reoviruses, Newcastle disease viruses, parvoviruses, measles viruses, vesicular stomatitis viruses (VSV), coxsackie viruses, and vaccinia viruses. In some embodiments, oncolytic viruses can specifically colonize solid tumors without infecting other organs and can be used as infectious agents to deliver nucleic acids encoding variant immunomodulatory polypeptides described herein to such solid tumors.

Oncolytic viruses for use in delivering nucleic acids encoding variant CD80 polypeptides or immunomodulatory proteins described herein can be any known to those of skill in the art, for example, vesicular stomatitis virus (Vs. stomatitis virus) (see, e.g., U.S. Pat. Nos. 7,731,974, 7,153,510, 6,653,103, and U.S. Patent Application Publication Nos. 2010/0178684, 2010/0172877, 2010/0113567, 2007/0098743, 20050260601, 20050220818, and European Patent Nos. 1385466, 1606411, and 1520175); herpes simplex virus (see, e.g., U.S. Pat. Nos. 7,897,146, 7,731,952, 7,5 Nos. 50,296, 7,537,924, 6,723,316, and 6,428,968, as well as U.S. Patent Application Publication Nos. 2014/0154216, 2011/0177032, 2011/0158948, 2010/0092515, 2009/0274728, 2009/0285860, 2009/0215147, 2009/0010889, 2007/0110720, 2006/0039894, 2004/0009604, and 2004/0063094, International Patent Publication WO 2007/052029, WO 1999/038955); retroviruses (see, e.g., U.S. Pat. Nos. 6,689,871, 6,635,472, 5,851,529, 5,716,826, 5,716,613, and U.S. Patent Application Publication No. 20110212530); vaccinia viruses (see, e.g., U.S. Pat. No. 2016/0339066), and adeno-associated viruses (see, e.g., U.S. Pat. No. 2016/0339066). (See Nos. 8,007,780, 7,968,340, 7,943,374, 7,906,111, 7,927,585, 7,811,814, 7,662,627, 7,241,447, 7,238,526, 7,172,893, 7,033,826, 7,001,765, 6,897,045, and 6,632,670).

Oncolytic viruses also include viruses that have been genetically modified to attenuate their pathogenicity, improve their safety profile, enhance their tumor specificity, and which also feature additional genes, such as cytotoxins, cytokines, prodrug-converting enzymes, that improve the overall efficacy of the virus (see, e.g., Kirn et al., (2009) Nat Rev Cancer 9:64-71; Garcia-Aragoncillo et al., (2010) Curr Opin Mol Ther 12:403-411; U.S. Patent Nos. 7,588,767, 7,588,771, 7,662,398, and 7,754,221, and U.S. Patent Application Publication Nos. 2007/0202572, 2007/0212727, 2010/0062016, 2009/00 (See, for example, Nos. 98529, 2009/0053244, 2009/0155287, 2009/0117034, 2010/0233078, 2009/0162288, 2010/0196325, 2009/0136917 and 2011/0064650). In some embodiments, the oncolytic virus can be a virus that is modified to selectively replicate in cancerous cells and is therefore oncolytic. For example, the oncolytic virus is an adenovirus that is modified to have modified tropism for tumor therapy and as a gene therapy vector. Examples include ONYX-015, H101 and Ad5ΔCR (Hallden and Portella (2012) Expert Opin Ther Targets, 16:945-58) and TNFerade (McLoughlin et al. (2005) Ann. Surg. Oncol., 12:825-30), or the conditionally replicating adenovirus Oncorine®.

In some embodiments, the infectious agent is a modified herpes simplex virus. In some embodiments, the infectious agent is a modified version of Talimogene laherparepvec (also known as T-Vec, Imlygic, or OncoVex GM-CSF) that has been modified to contain a nucleic acid encoding any of the variant immunomodulatory polypeptides described herein (e.g., any of the variant CD80 polypeptides or immunomodulatory proteins described herein). In some embodiments, the infectious agent is a modified herpes simplex virus, or variants thereof, as described, for example, in WO 2007/052029, WO 1999/038955, US 2004/0063094, US 2014/0154216.

In some embodiments, the infectious agent is a virus that targets a specific type of cell in the subject to which the virus is administered, e.g., a virus that targets immune cells or antigen-presenting cells (APCs). Dendritic cells (DCs) are APCs essential for initiating and controlling immune responses. DCs can capture and process antigens, migrate from the periphery to lymphoid organs, and present antigens to resting T cells in a major histocompatibility complex (MHC)-restricted manner. In some embodiments, the infectious agent is a virus that can specifically target DCs and deliver nucleic acids encoding variant CD80 polypeptides or immunomodulatory proteins for expression in DCs. In some embodiments, the virus is a lentivirus or a variant thereof. In some embodiments, the virus is a lentivirus pseudotyped to efficiently bind and productively infect cells, e.g., DCs, that express the cell surface marker dendritic cell-specific intercellular adhesion molecule-3-binding nonintegrin (DC-SIGN). In some embodiments, the virus is a lentivirus pseudotyped with the Sindbis virus E2 glycoprotein, such as those described in WO 2013/149167. In some embodiments, the virus allows for delivery and expression of a sequence of interest (e.g., a nucleic acid encoding any of the variant CD80 polypeptides or immunomodulatory proteins described herein) to DCs. In some embodiments, the virus includes those described in WO 2008/011636 or US 2011/0064763, or variants thereof.

2. Bacteria In some embodiments, the infectious agent is a bacterium.For example, in some embodiments, the bacterium can deliver any of the variant immunomodulatory polypeptides described herein, such as variant CD80 polypeptides or nucleic acids encoding immunomodulatory proteins, to target cells of a subject, such as tumor cells, immune cells, antigen-presenting cells and/or phagocytes.In some embodiments, the bacterium can be preferentially targeted to a specific environment within a subject, such as the tumor microenvironment (TME), for expression and/or secretion of the variant immunomodulatory polypeptide and/or for targeting specific cells of the environment for expression of the variant immunomodulatory polypeptide.

In some embodiments, the bacteria delivers the nucleic acid to the cell via bacterially-mediated transfer of plasmid DNA into mammalian cells (also referred to as "bactofection"). For example, in some embodiments, delivery of the genetic material is accomplished by invasion of the whole bacteria into the target cell. In some embodiments, spontaneous or induced bacteriolysis can lead to the release of the plasmid for subsequent eukaryotic expression. In some embodiments, the bacteria can deliver the nucleic acid to non-phagic mammalian cells (e.g., tumor cells) and/or phagocytes, e.g., certain immune cells and/or APCs. In some embodiments, the nucleic acid delivered by the bacteria can be transferred to the nucleus of the subject's cell for expression. In some embodiments, the nucleic acid also includes appropriate nucleic acid sequences, e.g., regulatory elements, e.g., promoters or enhancers, required for expression in a particular host cell of the operably linked sequence encoding the variant immunomodulatory polypeptide. In some embodiments, the infectious agent, a bacterium, can deliver the nucleic acid encoding the immunomodulatory protein in the form of RNA, e.g., pre-manufactured translation-competent RNA, that is delivered to the cytoplasm of the target cell for translation by the target cell's machinery.

In some embodiments, the bacteria can replicate in a target cell, e.g., a tumor cell, and lyse the cell. In some embodiments, the bacteria can contain and/or release a nucleic acid sequence and/or gene product in the cytoplasm of the target cell, thereby killing the target cell, e.g., a tumor cell. In some embodiments, the infectious agent is a bacterium that can replicate specifically in a particular environment of the subject, e.g., the tumor microenvironment (TME). For example, in some embodiments, the bacteria can replicate specifically in an anaerobic or hypoxic microenvironment. In some embodiments, conditions or factors present in a particular environment, e.g., aspartate, serine, citrate, ribose, or galactose produced by cells in the TME, can act as chemoattractants that attract the bacteria to the environment. In some embodiments, the bacteria can express and/or secrete an immunomodulatory protein as described herein in the environment, e.g., the TME.

In some embodiments, the infectious agent is a bacterium that is a Listeria sp., Bifidobacterium sp., Escherichia sp., Closteridium sp., Salmonella sp., Shigella sp., Vibrio sp., or Yersinia sp. In some embodiments, the bacteria is selected from the group consisting of Listeria monocytogenes, Salmonella typhimurium, Salmonella choleraesuis, Escherichia coli, Vibrio cholera, Clostridium perfringens, Clostridium butyricum, Clostridium novyi, Clostridium acetobutylicum, Bifidobacterium infantis, Bifidobacterium longum, and Bifidobacterium adolescentis. In some embodiments, the bacterium is selected from one or more of: In some embodiments, the bacteria is an engineered bacteria, e.g., Seow and Wood (2009) Molecular Therapy 17(5):767-777; Baban et al. (2010) Bioengineered Bugs 1:6, 385-394; Patyar et al. (2010) J Biomed Sci 17:21; Tangney et al. (2010) Bioengineered Bugs 1:4, 284-287; van Pijkeren et al. (2010) Hum Gene Ther. 21(4):405-416; WO 2012/149364; WO 2014/198002; US 9103831; US 9453227; US 2014/0186401; US 2004/0146488; 2011/0293705; US 2015/0359909 and EP 3020816. The bacteria can be modified to deliver nucleic acid sequences encoding any of the variant immunomodulatory polypeptides, conjugates and/or fusions provided herein and/or to express such variant immunomodulatory polypeptides in a subject.

F. Nucleic Acids, Vectors, and Methods or Cells for Producing Polypeptides Provided herein are isolated or recombinant nucleic acids (collectively "nucleic acids") that code for any of the various provided aspects of the variant CD80 polypeptides or immunomodulatory polypeptides provided herein. In some embodiments, the nucleic acids provided herein, including all of the following, are useful for recombinant production (e.g., expression) of the variant CD80 polypeptides or immunomodulatory polypeptides provided herein. In some embodiments, the nucleic acids provided herein, including all of the following, are useful for expression in cells, e.g., modified cells, e.g., immune cells, or infectious agent cells, of the variant CD80 polypeptides or immunomodulatory polypeptides provided herein. The nucleic acids provided herein can be in the form of RNA or DNA, and can include mRNA, cRNA, recombinant or synthetic RNA and DNA, as well as cDNA. The nucleic acids provided herein are typically DNA molecules, usually double-stranded DNA molecules. However, single-stranded DNA, single-stranded RNA, double-stranded RNA, and hybrid DNA/RNA nucleic acids or combinations thereof, comprising any of the nucleotide sequences of the present invention, are also provided.

Also provided herein are recombinant expression vectors and recombinant host cells useful in producing the variant CD80 polypeptides or immunomodulatory polypeptides provided herein.

Also provided herein are modified cells (e.g., modified immune cells) that contain any of the provided nucleic acid molecules or the encoded variant CD80 polypeptides or immunomodulatory polypeptides (e.g., any of the transmembrane immunomodulatory polypeptides or secreted immunomodulatory polypeptides).

Also provided herein are infectious agents (e.g., bacterial or viral cells) that contain any of the provided nucleic acid molecules or the encoded variant CD80 polypeptides or immunomodulatory polypeptides (e.g., any of the transmembrane immunomodulatory polypeptides or secreted immunomodulatory polypeptides).

In any of the embodiments provided above, a nucleic acid encoding an immunomodulatory polypeptide provided herein can be introduced into a cell using recombinant DNA and cloning techniques. To do so, a recombinant DNA molecule encoding the immunomodulatory polypeptide is prepared. Methods for preparing such DNA molecules are well known in the art. For example, the peptide-encoding sequence can be excised from the DNA using a suitable restriction enzyme. Alternatively, the DNA molecule can be synthesized using chemical synthesis techniques such as the phosphoramidite method. A combination of these techniques can also be used. In some cases, the recombinant or synthetic nucleic acid can be generated by polymerase chain reaction (PCR). In some embodiments, a DNA insert encoding one or more variant CD80 polypeptides containing at least one affinity-modified IgSF domain and, in some embodiments, a signal peptide, a transmembrane domain, and/or an endodomain can be generated according to the provided instructions. The DNA insert can be cloned into an appropriate transduction/transfection vector as known to those skilled in the art. Also provided is an expression vector containing the nucleic acid molecule.

In some embodiments, the expression vector is capable of expressing an immunomodulatory protein in a suitable cell under conditions suitable for expression of the protein. In some aspects, the nucleic acid molecule or expression vector comprises a DNA molecule encoding an immunomodulatory protein operably linked to an appropriate expression control sequence. Methods for achieving this functional linkage either before or after the DNA molecule is inserted into the vector are well known. Expression control sequences include promoters, activators, enhancers, operators, ribosome binding sites, start signals, stop signals, cap signals, polyadenylation signals, and other signals involved in the control of transcription or translation.

In some embodiments, expression of the immunomodulatory protein is controlled by a promoter or enhancer to control or regulate expression. The promoter is operably linked to a portion of the nucleic acid molecule encoding the variant polypeptide or the immunomodulatory protein. In some embodiments, the promoter is a constitutively active promoter (e.g., a tissue-specific constitutively active promoter or other constitutive promoter). In some embodiments, the promoter is an inducible promoter that can be responsive to an inducer (e.g., a T cell activation signal).

In some embodiments, a constitutive promoter is operably linked to a nucleic acid molecule encoding a variant polypeptide or an immunomodulatory protein. Exemplary constitutive promoters include the simian vacuolar virus 40 (SV40) promoter, the cytomegalovirus (CMV) promoter, the ubiquitin C (UbC) promoter, and the EF-1 alpha (EF1a) promoter. In some embodiments, the constitutive promoter is tissue specific. For example, in some embodiments, the promoter allows for constitutive expression of the immunomodulatory protein in a particular tissue, such as an immune cell, lymphocyte, or T cell. Exemplary tissue-specific promoters are described in U.S. Pat. No. 5,998,205, including, for example, the fetoprotein, DF3, tyrosinase, CEA, surfactant protein, and ErbB2 promoters.

In some embodiments, an inducible promoter is operably linked to a nucleic acid molecule encoding a variant polypeptide or an immunomodulatory protein such that expression of the nucleic acid is controllable by controlling the presence or absence of an appropriate inducer of transcription. For example, the promoter can be a regulated promoter and transcription factor expression system, such as the published tetracycline-regulated system or other regulatable systems (see, for example, published International PCT application WO 01/30843), that allows for regulated expression of the encoded polypeptide. An exemplary regulatable promoter system is the Tet-On (and Tet-Off) system, available, for example, from Clontech (Palo Alto, CA). This promoter system allows for regulated expression of the transgene, controlled by tetracycline or a tetracycline derivative, such as doxycycline. Other regulatable promoter systems are known (see, e.g., published U.S. Patent Application No. 2002-0168714, entitled "Regulation of Gene Expression Using Single-Chain, Monomeric, Ligand Dependent Polypeptide Switches," which describes gene switches containing ligand binding domains and transcriptional regulatory domains, e.g., those derived from hormone receptors).

In some embodiments, the promoter is responsive to an element responsive to T cell activation signaling. As an example, in some embodiments, the modified T cell comprises an expression vector encoding an immunomodulatory protein and a promoter operably linked to regulated expression of the immunomodulatory protein. The modified T cell can be activated, for example, by signaling through a modified T cell receptor (TCR) or chimeric antigen receptor (CAR), thereby causing expression and secretion of the immunomodulatory protein by the responsive promoter.

In some embodiments, an inducible promoter is operably linked to a nucleic acid molecule encoding an immunomodulatory protein such that the immunomodulatory protein is expressed in response to nuclear factor of activated T cells (NFAT) or nuclear factor kappa light chain enhancer of activated B cells (NF-κB). For example, in some embodiments, the inducible promoter comprises a binding site for NFAT or NF-κB. For example, in some embodiments, the promoter is an NFAT or NF-κB promoter, or a functional variant thereof. Thus, in some embodiments, the nucleic acid allows for controlling expression of the immunomodulatory protein while also reducing or eliminating toxicity of the immunomodulatory protein. In particular, modified immune cells comprising the nucleic acid of the invention express and secrete the immunomodulatory protein only when the cell (e.g., a cell expressing a T cell receptor (TCR) or a chimeric antigen receptor (CAR)) is specifically stimulated by an antigen and/or when the cell (e.g., the calcium signaling pathway of the cell) is non-specifically stimulated, for example, by phorbol myristate acetate (PMA)/ionomycin. Thus, expression, and in some cases secretion, of immunomodulatory proteins can be controlled to occur only when and where it is needed (e.g., in the presence of an infectious agent, cancer, or at a tumor site), thus reducing or avoiding undesirable immunomodulatory protein interactions.

In some embodiments, the nucleic acid encoding the immunomodulatory protein described herein comprises a suitable nucleotide sequence encoding a NFAT promoter, a NF κB promoter, or a functional variant thereof. "NFAT promoter" as used herein means one or more NFAT responsive elements linked to a minimal promoter. "NF κB promoter" refers to one or more NF κB responsive elements linked to a minimal promoter. In some embodiments, the minimal promoter of a gene is a minimal human IL-2 promoter or a CMV promoter. The NFAT responsive element may include, for example, NFAT1, NFAT2, NFAT3, and/or NFAT4 responsive elements. The NFAT promoter, NF κB promoter, or a functional variant thereof may include any number of binding motifs, for example, at least 2, at least 3, at least 4, at least 5, or at least 6, at least 7, at least 8, at least 9, at least 10, at least 11, or up to 12 binding motifs.

The resulting recombinant expression vector having the DNA molecule thereon is used to transform a suitable host. This transformation can be performed using methods well known in the art. In some embodiments, the nucleic acid provided herein further comprises a nucleotide sequence encoding a secretory or signal peptide operably linked to the nucleic acid encoding the immunomodulatory polypeptide such that the resulting soluble immunomodulatory polypeptide is recovered from the culture medium, the host cell, or the host cell periplasm. In other embodiments, appropriate expression control signals are selected to allow membrane expression of the immunomodulatory polypeptide. Additionally, commercially available kits and contract manufacturers can be utilized to produce the modified cells or recombinant host cells provided herein.

In some embodiments, the resulting expression vector having the DNA molecule thereon is used to transform, e.g., transduce, a suitable cell. The introduction can be performed using methods well known in the art. Exemplary methods include those for the transfer of the nucleic acid encoding the receptor, including via a virus, e.g., a retrovirus or lentivirus, transduction, transposon, and electroporation. In some embodiments, the expression vector is a viral vector. In some embodiments, the nucleic acid is transferred to the cell by lentiviral or retroviral transduction methods.

Any of a number of publicly available and well-known mammalian host cells, including mammalian T cells or APCs, can be used in preparing the polypeptides or modified cells. The choice of cell will depend on a number of factors recognized by the art. These include, for example, compatibility with the selected expression vector, toxicity of the peptide encoded by the DNA molecule, transformation rate, ease of recovery of the peptide, expression characteristics, biosafety, and cost. A balance of these factors must be made with the understanding that not all cells will be equally effective in expressing a particular DNA sequence.

In some embodiments, the host cells can be a wide variety of eukaryotic cells, such as yeast cells, or mammalian cells, such as Chinese hamster ovary (CHO) or HEK293 cells. The host cells can also be prokaryotic cells, such as Escherichia coli (E. coli). The transformed recombinant host is cultured under conditions to express the polypeptide, which is then purified to obtain soluble protein. The recombinant host cells can be cultured under conventional fermentation conditions to express the desired polypeptide. Such fermentation conditions are well known in the art. Finally, the polypeptides provided herein can be recovered and purified from the recombinant cell culture by any of a number of methods well known in the art, including ammonium sulfate or ethanol precipitation, acid extraction, anion or cation exchange chromatography, cellulose phosphate chromatography, hydrophobic interaction chromatography, and affinity chromatography. If desired, a protein refolding step can be used in completing the configuration of the mature protein. Finally, high performance liquid chromatography (HPLC) can be employed for the final purification step.

In some embodiments, the cells are any of those described above in connection with preparing immune cells, e.g., modified cells. In some embodiments, such modified cells are primary cells. In some embodiments, the modified cells are autologous to the subject. In some embodiments, the modified cells are allogeneic to the subject. In some embodiments, the modified cells are obtained from the subject, e.g., by leukapheresis, and transformed ex vivo for expression of an immunomodulatory polypeptide, e.g., a transmembrane immunomodulatory polypeptide or a secreted immunomodulatory polypeptide.

Also provided are nucleic acids encoding any of the variant immunomodulatory polypeptides contained in the infectious agents described herein. In some embodiments, the infectious agents deliver the nucleic acid to cells of a subject and/or permit expression in the cells of the encoded variant polypeptide. Also provided are nucleic acids used to generate, produce, or regulate such infectious agents. For example, in some embodiments, vectors and/or plasmids are provided that contain nucleic acids encoding variant immunomodulatory polypeptides for generation of infectious agents, delivery to cells of a subject, and/or expression in cells of a subject of the variant immunomodulatory polypeptide.

In some embodiments, the nucleic acid provided is a recombinant viral or bacterial vector containing a nucleic acid sequence encoding a variant immunomodulatory polypeptide. In some embodiments, the recombinant vector can be used to produce and/or deliver to a target cell of a subject an infectious agent containing a nucleic acid sequence encoding a variant immunomodulatory polypeptide for expression by the target cell. In some embodiments, the recombinant vector is an expression vector. In some embodiments, the recombinant vector contains appropriate sequences necessary for the generation and/or production of an infectious agent and expression in a target cell.

In some embodiments, the recombinant vector is a plasmid or cosmid. Plasmids or cosmids containing a nucleic acid sequence encoding a variant immunomodulatory polypeptide as described herein are readily constructed using standard techniques well known in the art. For production of infectious agents, the vector or genome can be constructed in a plasmid form, which can then be transfected into a packaging or production cell line or host bacterium. Recombinant vectors can be generated using any of the recombinant techniques known in the art. In some embodiments, the vector can include a prokaryotic origin of replication and/or a gene whose expression is detectable or selectable marker, e.g., a gene that confers drug resistance for propagation and/or selection in a prokaryotic system.

In some embodiments, the recombinant vector is a viral vector. Exemplary recombinant viral vectors include lentivirus vector genomes, poxvirus vector genomes, vaccinia virus vector genomes, adenovirus vector genomes, adenovirus-associated virus vector genomes, herpesvirus vector genomes, and alphavirus vector genomes. The viral vector can be a live viral vector, an attenuated viral vector, a replication-conditional or replication-deficient viral vector, a non-pathogenic (defective) viral vector, a replication-competent viral vector, and/or modified to express a heterologous gene product, e.g., a variant immunomodulatory polypeptide provided herein. Vectors for the generation of viruses can also be modified to alter the attenuation of the virus, including any method that increases or decreases the transcriptional or translational load.

Exemplary viral vectors that can be used include modified vaccinia virus vectors (see, e.g., Guerra et al., J. Virol. 80:985-98 (2006); Tartaglia et al., AIDS Research and Human Retroviruses 8: 1445-47 (1992); Gheradi et al., J. Gen. Virol. 86:2925-36 (2005); Mayr et al., Infection 3:6-14 (1975); Hu et al., J. Virol. 75: 10300-308 (2001); U.S. Pat. Nos. 5,698,530, 6,998,252, 5,443,964, 7,247,615, and 7,368,116); adenovirus vectors or adenovirus-associated virus vectors (see, e.g., Molin et al., J. Virol. 80:985-98 (2006); Tartaglia et al., AIDS Research and Human Retroviruses 8: 1445-47 (1992); Gheradi et al., J. Gen. Virol. 86:2925-36 (2005); Mayr et al., Infection 3:6-14 (1975); Hu et al., J. Virol. 75: 10300-308 (2001); U.S. Pat. Nos. 5,698,530, 6,998,252, 5,443,964, 7,247,615, and 7,368,116). al., J. Virol. 72:8358-61 (1998); Narumi et al., Am J. Respir. Cell Mol. Biol. 19:936-41 (1998); Mercier et al., Proc. Natl. Acad. Sci. USA 101:6188-93 (2004); U.S. Patent Nos. 6,143,290; 6,596,535; 6,855,317; 6,936,257; 7,125,717; 7,378,087; and 7,550,296); retroviral vectors, including those based on murine leukemia virus (MuLV), gibbon ape leukemia virus (GaLV), ecotropic retroviruses, simian immunodeficiency virus (SIV), human immunodeficiency virus (HIV), and combinations (see, e.g., Buchscher et al., J. Virol. 66:2731-39 (1992); Johann et al., J. Virol. 66: 1635-40 (1992); Sommerfelt et al., Virology 176:58-59 (1990); Wilson et al., J. Virol. 63:2374-78 (1989); Miller et al., J. Virol. 65:2220-24 (1991); Miller et al., Mol. Cell Biol. 10:4239 (1990); Kolberg, NIH Res. 4:43 1992; Cornetta et al., Hum. Gene Ther. 2:215 (1991) , 2006; see, e.g., 1999; Fitzgerald et al., J. Virol. 18:131-132, 2007; Ford et al., J. Virol. 19 ... 73:9011 (1999); WO 2009/076524; WO 2012/141984; WO 2016/011083; McWilliams et al., J. Virol. 77: 11150, 2003; Powell et al., J. Virol. 70:5288, 1996) or any variants thereof, and/or vectors that can be used to generate any of the above viruses. In some embodiments, the recombinant vector can include a regulatory sequence, such as a promoter or enhancer sequence, that can regulate expression of the viral genome, e.g., in the case of an RNA virus, in a packaging cell line (see, e.g., U.S. Patent Nos. 5,385,839 and 5,168,062).

In some embodiments, the recombinant vector is an expression vector, e.g., an expression vector that permits expression of an encoded gene product when delivered to a target cell, e.g., a cell of a subject, e.g., a tumor cell, an immune cell, and/or an APC. In some embodiments, the recombinant expression vector contained in the infectious agent is capable of expressing an immunomodulatory protein in a target cell of the subject under conditions suitable for expression of the protein.

In some aspects, the nucleic acid or expression vector comprises a nucleic acid sequence encoding an immunomodulatory protein operably linked to an appropriate expression control sequence. Methods for effecting this functional linkage either before or after the nucleic acid sequence encoding the immunomodulatory protein is inserted into the vector are well known. Expression control sequences include promoters, activators, enhancers, operators, ribosome binding sites, start signals, stop signals, cap signals, polyadenylation signals, and other signals involved in the control of transcription or translation. The promoter can be operably linked to a portion of the nucleic acid sequence encoding the immunomodulatory protein. In some embodiments, the promoter is a constitutively active promoter in the target cell (e.g., a tissue-specific constitutively active promoter or other constitutive promoter). For example, the recombinant expression vector can also include a lymphoid tissue-specific transcriptional regulatory element (TRE), such as a B-lymphocyte, T-lymphocyte, or dendritic cell-specific TRE. Lymphoid tissue-specific TREs are known in the art (see, e.g., Thompson et al., Mol. Cell. Biol. 12:1043-53 (1992); Todd et al., J. Exp. Med. 177:1663-74 (1993); Penix et al., J. Exp. Med. 178:1483-96 (1993)). In some embodiments, the promoter is an inducible promoter that can be responsive to an inducer (e.g., a T cell activation signal). In some embodiments, the nucleic acid delivered to a target cell of interest, e.g., a tumor cell, an immune cell, and/or an APC, can be operably linked to any of the above regulatory elements.

In some embodiments, the vector is a bacterial vector, e.g., a bacterial plasmid or cosmid. In some embodiments, the bacterial vector is delivered to a target cell, e.g., a tumor cell, an immune cell, and/or an APC, via bacterial-mediated transfer of plasmid DNA into a mammalian cell (also referred to as "bactofection"). In some embodiments, the delivered bacterial vector also contains appropriate expression control sequences for expression in the target cell, e.g., a promoter sequence and/or an enhancer sequence, or any of the regulatory or control sequences described above. In some embodiments, the bacterial vector contains appropriate expression control sequences for expression and/or secretion of the encoded variant polypeptide in an infectious agent, e.g., a bacterium.

In some embodiments, the polypeptides provided herein can also be produced by synthetic methods. Solid-phase synthesis is the preferred technique for producing individual peptides because it is the most cost-effective method for producing small peptides. For example, well-known solid-phase synthesis techniques include the use of protecting groups, linkers, and solid-phase supports, as well as specific protection and deprotection reaction conditions, linker cleavage conditions, the use of scavengers, and other aspects of solid-phase peptide synthesis. The peptides can then be assembled into polypeptides as provided herein.

IV. METHODS FOR ASSESSING THE IMMUNOMODULATORY ACTIVITY OF VARIANT CD80 POLYPEPTIDES AND IMMUNOMODULATORY PROTEINS
In some embodiments, the variant CD80 polypeptides provided herein (e.g., full-length and/or specific binding fragments or conjugates, stack constructs or derivatives thereof) exhibit immunomodulatory activity that modulates T cell activation. In some embodiments, the CD80 polypeptides modulate IFN-γ expression in primary T cell assays compared to wild-type or unmodified CD80 controls. In some cases, the modulation of IFN-γ expression can increase or decrease IFN-γ expression compared to controls. Assays to determine specific binding and IFN-γ expression are well known in the art and include the MLR (mixed lymphocyte reaction) assay measuring interferon-γ cytokine levels in culture supernatants (Wang et al., Cancer Immunol Res. 2014 Sep: 2(9):846-56), the SEB (Staphylococcal enterotoxin B) T cell stimulation assay (Wang et al., Cancer Immunol Res. 2014 Sep: 2(9):846-56), and the anti-CD3 T cell stimulation assay (Li and Kurlander, J Transl Med. 2010: 8: 104).

In some embodiments, the variant CD80 polypeptides can increase, or in alternative embodiments decrease, IFN-γ (interferon-γ) expression in primary T cell assays relative to a wild-type CD80 control. In some embodiments of the provided polypeptides containing a soluble variant CD80 sequence, the polypeptides can increase, or in alternative embodiments decrease, IFN-γ expression in primary T cell assays relative to a wild-type CD80 control. In some embodiments of the provided polypeptides containing multiple variant CD80 sequences, the polypeptides can increase, or in alternative embodiments decrease, IFN-γ expression in primary T cell assays relative to a wild-type CD80 control.

One of skill in the art will recognize that the format of the primary T cell assay used to determine increased IFN-γ expression can differ from that employed to assay decreased IFN-γ expression. In assaying the ability of a variant CD80 to decrease IFN-γ expression in a primary T cell assay, a mixed lymphocyte reaction (MLR) assay can be used as described in Example 6. In some cases, a soluble form of the variant CD80 can be employed to determine the ability of the variant CD80 to antagonize and thereby decrease IFN-γ expression in an MLR, as described in Example 6.

Alternatively, in assaying the ability of variant CD80 to increase IFN-γ expression in primary T cell assays, a co-fixation assay can be used as described in Example 6. In the co-fixation assay, a TCR signal (provided in some embodiments by an anti-CD3 antibody) is used in conjunction with co-fixed variant CD80 to determine the ability to increase IFN-γ expression relative to a CD80 control. In some cases, a soluble form of variant CD80 multimerized to a degree that provides multivalent binding can be employed to determine the ability of the variant CD80 to turn on IFN-γ expression in the MLR, thereby increasing its expression, as described in Example 6.

The use of suitable controls will be known to those of skill in the art, but in the foregoing embodiments, the control typically involves the use of unmodified CD80, e.g., a wild-type, native CD80 isoform from the same mammalian species from which the variant CD80 is derived or developed. Regardless of whether it has increased or decreased binding affinity to one or both of ICOS and CD28, the variant CD80 increases IFN-γ expression in some embodiments, and decreases IFN-γ expression in alternative embodiments, relative to a wild-type CD80 control in primary T cell assays.

In some embodiments, the variant CD80 increases IFN-γ expression (i.e., protein expression) by at least 5%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90% or more compared to a wild-type or unmodified CD80 control. In other embodiments, the variant CD80 decreases IFN-γ expression (i.e., protein expression) by at least 5%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90% or more compared to a wild-type or unmodified CD80 control. In some embodiments, the wild-type CD80 control is murine CD80, as typically used for variant CD80s that have a sequence change from the wild-type murine CD80 sequence. In some embodiments, the wild-type CD80 control is human CD80, as typically used for variant CD80s that have a sequence that is altered from a wild-type human CD80 sequence, e.g., a CD80 sequence that includes the amino acid sequence of SEQ ID NO:28, SEQ ID NO:152, or SEQ ID NO:372.

V. Pharmaceutical Formulations
Compositions containing any of the variant CD80 polypeptides, immunomodulatory proteins, conjugates, modified cells or infectious agents described herein are provided herein. Pharmaceutical compositions can further include pharma- ceutical acceptable excipients. For example, pharmaceutical compositions can contain one or more excipients for adjusting, maintaining, or preserving, for example, pH, osmolality, viscosity, transparency, color, isotonicity, odor, sterility, stability, dissolution or release rate, adsorption, or penetration of the composition. In some aspects, those skilled in the art will understand that pharmaceutical compositions containing cells may be different from pharmaceutical compositions containing proteins.

In some embodiments, the pharmaceutical composition is a solid, e.g., a powder, capsule, or tablet. For example, the components of the pharmaceutical composition can be lyophilized. In some embodiments, the solid pharmaceutical composition is reconstituted or dissolved in a liquid prior to administration.

In some embodiments, the pharmaceutical composition is a variant CD80 polypeptide dissolved in a liquid, e.g., an aqueous solution (e.g., saline or Ringer's solution). In some embodiments, the pH of the pharmaceutical composition is about 4.0 to about 8.5 (e.g., about 4.0 to about 5.0, about 4.5 to about 5.5, about 5.0 to about 6.0, about 5.5 to about 6.5, about 6.0 to about 7.0, about 6.5 to about 7.5, about 7.0 to about 8.0, or about 7.5 to about 8.5).

In some embodiments, the pharmaceutical composition comprises a pharma- ceutically acceptable excipient, such as a filler, binder, coating agent, preservative, lubricant, flavoring agent, sweetener, coloring agent, solvent, buffer, chelating agent, or stabilizer. Examples of pharma-ceutically acceptable fillers include cellulose, dicalcium phosphate, calcium carbonate, microcrystalline cellulose, sucrose, lactose, glucose, mannitol, sorbitol, maltol, pregelatinized starch, corn starch, or potato starch. Examples of pharma-ceutically acceptable binders include polyvinylpyrrolidone, starch, lactose, xylitol, sorbitol, maltitol, gelatin, sucrose, polyethylene glycol, methylcellulose, or cellulose. Examples of pharma-ceutically acceptable coating agents include hydroxypropylmethylcellulose (HPMC), shellac, corn protein zein, or gelatin. Examples of pharma-ceutically acceptable disintegrants include polyvinylpyrrolidone, carboxymethylcellulose, or sodium starch glycolate. Examples of pharma- ceutically acceptable lubricants include polyethylene glycol, magnesium stearate, or stearic acid. Examples of pharma-ceutically acceptable preservatives include methylparaben, ethylparaben, propylparaben, benzoic acid, or sorbic acid. Examples of pharma-ceutically acceptable sweeteners include sucrose, saccharin, aspartame, or sorbitol. Examples of pharma-ceutically acceptable buffers include carbonates, citrates, gluconates, acetates, phosphates, or tartrates.

In some embodiments, the pharmaceutical composition further comprises a material for controlled or sustained release of the product, such as injectable microspheres, bioerodible particles, polymeric compounds (polylactic acid, polyglycolic acid), beads, or liposomes.

In some embodiments, the pharmaceutical compositions are sterile. Sterilization may be accomplished by filtration through sterile filtration membranes or by irradiation. If the composition is lyophilized, sterilization using this method may be performed prior to or after lyophilization and reconstitution. Compositions for parenteral administration may be stored in lyophilized form or in a solution. In addition, parenteral compositions generally are placed into a container having a sterile access port, for example, an intravenous solution bag or vial having a stopper pierceable by a hypodermic injection needle.

In some embodiments, pharmaceutical compositions containing transmembrane immunomodulatory proteins are provided, including modified cells expressing such transmembrane immunomodulatory proteins. In some embodiments, pharmaceutical compositions and formulations include one or more any pharma- ceutically acceptable carriers or excipients. Such compositions may include a buffer, such as neutral buffered saline, phosphate buffered saline, etc.; a carbohydrate, such as glucose, mannose, sucrose or dextran, mannitol; a protein; a polypeptide or an amino acid, such as glycine; an antioxidant; a chelating agent, such as EDTA or glutathione; an adjuvant (e.g., aluminum hydroxide); and a preservative. The compositions of the invention are preferably formulated for intravenous administration.

In some embodiments, the pharmaceutical composition contains an infectious agent that contains a nucleic acid sequence encoding an immunomodulatory variant polypeptide. In some embodiments, the pharmaceutical composition contains a dose of the infectious agent suitable for administration to a subject suitable for treatment. In some embodiments, the pharmaceutical composition contains a single or multiple dose of an infectious agent that is a virus, e.g., at ^least , or at least about, 1x10, ^1x10 , ^1x10 , 1x10, ^2x10 , ^3x10 , 4x10, ^5x10 pfu or about ^1x10 pfu, including ^{1x10 to 1x10} plaque forming units ⁽ pfu), ^1x10 to ^{1x10 pfu} or about 1x10 to ^{1x10 pfu} , or 1x10 to ^{1x10 pfu or} about ^{1x10 to 1x10 pfu} , respectively. In some embodiments, the pharmaceutical composition can contain a virus concentration of ¹⁰ to ¹⁰ or about 10 to about ¹⁰ pfu/mL, e.g., ^5x10 to ^5x10 or about 5x10 to about ^5x10 , or ^1x10 to ^1x10 or ^{about 1x10} to about ^1x10 pfu/mL, e.g., at least, or at least about, 10 pfu/mL, ¹⁰ pfu ^/ mL, ¹⁰ pfu/mL, ¹⁰ pfu/mL or ¹⁰ pfu/mL. In some embodiments, the pharmaceutical composition contains a bacterial infectious agent in single or multiple doses of, e.g., ^{at least} , or at least about, 1x104, 1x105, 1x106, 1x107, ^{1x108 or 1x109 cfu, including 1x103 to 1x109 or about 1x103 to about 1x109 colony forming units (cfu), 1x104 to 1x109 cfu or about 1x104 to about 1x109 cfu} , or ^{1x105 to 1x107 cfu} or about ^1x105 to ^{about 1x107} cfu, respectively. In some embodiments, the pharmaceutical composition can contain a bacterial concentration of ¹⁰ to ¹⁰ or about 10 to about ¹⁰ cfu/mL, e.g., ^{5x10 to 5x10} or about ^5x10 to about 5x10, or ^1x10 to ^1x10 or ^{about 1x10} to about ^1x10 cfu/mL, e.g., at least, or at least about, ¹⁰ cfu/mL, ¹⁰ cfu/mL, ¹⁰ cfu/mL or ¹⁰ cfu/mL.

Such a formulation may be, for example, in a form suitable for intravenous infusion. A pharma- ceutically acceptable carrier may be any pharma- ceutically acceptable material, composition, or vehicle involved in the carrying or transport of the cells of interest from one tissue, organ, or part of the body to another. For example, the carrier may be a liquid or solid filler, diluent, excipient, solvent, or encapsulating material, or some combination thereof. Each component of the carrier must be "pharma- ceutically acceptable" in that it must be compatible with the other active ingredients of the formulation. It must also be suitable for contact with any tissue, organ, or part of the body that may be encountered, in the sense that it should not have a risk of toxicity, irritation, allergic response, immunogenicity, or any other complication that unduly outweighs its therapeutic benefits.

In some embodiments, the pharmaceutical composition is administered to a subject. In general, the dosage and route of administration of the pharmaceutical composition are determined according to standard pharmaceutical practice, depending on the size and condition of the subject. For example, a therapeutically effective dose can be estimated initially, either in cell culture assays or in animal models, such as mice, rats, rabbits, dogs, pigs, or monkeys. Animal models may also be used to determine appropriate concentration ranges and routes of administration. Such information can then be used to determine useful doses and routes of administration in humans. The exact dosage will be determined in light of factors related to the treatment required by the subject. Dosage and administration are adjusted to provide sufficient levels of the active compound or to maintain the desired effect. Factors that may be considered include the severity of the disease state, the overall health of the subject, the age, weight, and sex of the subject, the time and frequency of administration, drug combinations, reaction sensitivities, and response to treatment.

Long-acting pharmaceutical compositions may be administered every 3-4 days, weekly, or every other week, depending on the half-life and clearance rate of the particular formulation. The frequency of dosing will depend on the pharmacokinetic parameters of the molecule in the formulation used. Typically, the composition is administered until a dosage is reached that achieves the desired effect. Thus, the composition may be administered as a single dose or as multiple doses (at the same or different concentrations/dosages) over time, or as a continuous infusion. Further refinement of the appropriate dosage is routinely performed. The appropriate dosage may be ascertained through the use of appropriate dose-response data. A number of biomarkers or physiological markers of therapeutic efficacy may be monitored, including T cell activation or proliferation, cytokine synthesis or production (e.g., production of TNF-α, IFN-γ, IL-2), induction of various activation markers (e.g., CD25, IL-2 receptor), inflammation, joint swelling or tenderness, serum levels of C-reactive protein, anti-collagen antibody production, and/or T cell-dependent antibody responses.

In some embodiments, the pharmaceutical composition is administered to a subject through any route, including orally, transdermally, by inhalation, intravenously, intraarterially, intramuscularly, directly at a wound site, at a surgical site, intraperitoneally, by suppository, subcutaneously, intradermally, transdermally, by spray, intrapleurally, intraventricularly, intraarticularly, intraocularly, or intrathecally.

In some embodiments, the administration of the pharmaceutical composition is a single dose or multiple doses. In some embodiments, the pharmaceutical composition is administered to the subject once a day, twice a day, three times a day, or four or more times a day. In some embodiments, about one or more doses (e.g., about two or more doses, about three or more doses, about four or more doses, about five or more doses, about six or more doses, or about seven or more doses) are given per week. In some embodiments, multiple doses are given over the course of days, weeks, months, or years. In some embodiments, the course of treatment is about one or more doses (e.g., about two or more doses, about three or more doses, about four or more doses, about five or more doses, about seven or more doses, about ten or more doses, about fifteen or more doses, about twenty-five or more doses, about forty or more doses, about fifty or more doses, or about one hundred or more doses).

In some embodiments, the dose of the pharmaceutical composition administered is about 1 μg or more of protein per kg of subject body weight (e.g., about 2 μg or more of protein per kg of subject body weight, about 5 μg or more of protein per kg of subject body weight, about 10 μg or more of protein per kg of subject body weight, about 25 μg or more of protein per kg of subject body weight, about 50 μg or more of protein per kg of subject body weight, about 100 μg or more of protein per kg of subject body weight, about 250 μg or more of protein per kg of subject body weight, about 500 μg or more of protein per kg of subject body weight, about 1 mg or more of protein per kg of subject body weight, about 2 mg or more of protein per kg of subject body weight, or about 5 mg or more of protein per kg of subject body weight).

In some embodiments, a therapeutic amount of the cell composition is administered. Typically, the exact amount of the composition of the invention to be administered can be determined by a physician taking into account individual differences in the age, weight, tumor size, extent of infection or metastasis, and condition of the patient (subject). In general, it can be provided that a pharmaceutical composition comprising modified cells, e.g., T cells, as described herein, can be administered at a dosage of ¹⁰⁴ to ¹⁰⁹ cells/kg body weight, e.g., ¹⁰⁵ to ¹⁰⁶ cells/kg body weight, including all integer values within these ranges. Also, the modified cell composition, e.g., T cell composition, can be administered multiple times at these dosages. The cells can be administered by using injection techniques commonly known in immunotherapy (see, e.g., Rosenberg et al, New Eng. J. of Med. 319: 1676, 1988). Optimal dosages and treatment regimes for a particular patient can be readily determined by one skilled in the medical arts by monitoring the patient's disease symptoms and adjusting the treatment accordingly.

A wide variety of means are known for determining whether administration of the therapeutic compositions of the invention sufficiently modulates immune activity, including by eliminating, sequestering, or inactivating immune cells that mediate or are capable of mediating an undesirable immune response; inducing, generating, or stimulating immune cells that mediate or are capable of mediating a protective immune response; altering the physical or functional properties of immune cells; or a combination of these effects. Examples of measurements of modulation of immune activity include, but are not limited to, testing for the presence or absence of immune cell populations (using flow cytometry, immunohistochemistry, histology, electron microscopy, polymerase chain reaction (PCR)); measuring the functional capacity of immune cells, including their ability to proliferate or differentiate in response to a signal, or their resistance to proliferation or differentiation (e.g., T cell proliferation assays and 3H-thymidine following stimulation with anti-CD3 antibodies, anti-T cell receptor antibodies, anti-CD28 antibodies, calcium ionophores, PMA (phorbol 12-myristate 13-acetate), antigen-presenting cells loaded with peptide or protein antigens. uptake-based pepscan analysis; using B cell proliferation assays); measuring the ability to kill or lyse other cells (e.g., cytotoxic T cell assays); measuring cytokines, chemokines, cell surface molecules, antibodies and other products of cells (e.g., by flow cytometry, enzyme-linked immunosorbent assay, Western blot analysis, protein microarray analysis, immunoprecipitation analysis); measuring biochemical markers of activation of immune cells or signaling pathways within immune cells (e.g., tyrosine, serine or threonine phosphorylation, polypeptide cleavage, and protein complex formation or dissociation). Western blot and immunoprecipitation analysis of ; protein array analysis; DNA array or DNA transcription profiling using subtractive hybridization); measurement of cell death by apoptosis, necrosis or other mechanisms (e.g., Annexin V staining, TUNEL assay, gel electrophoresis measuring DNA laddering, histology; fluorogenic caspase assay, Western blot analysis of caspase substrates); measurement of genes, proteins, and other molecules produced by immune cells (e.g., Northern blot analysis, polymerase chain reaction, DNA microarrays, protein microarrays, two-dimensional gel electrophoresis, Western blot analysis, enzyme-linked immunosorbent assay, flow cytometry); and measurement (clinical scores, requirement to use additional therapy, functional status, imaging studies) of clinical symptoms or clinical outcomes (e.g., improvement) of autoimmune diseases, neurodegenerative diseases, and other diseases involving self-proteins or self-polypeptides, for example, by measuring relapse rates or disease severity in the case of multiple sclerosis (using clinical scores known to those of skill in the art), blood glucose in the case of type I diabetes, or joint inflammation in the case of rheumatoid arthritis.

VI. Manufactured Articles and Kits
Also provided herein is an article of manufacture that includes the pharmaceutical composition described herein in suitable packaging.Suitable packaging for the compositions described herein (e.g., ophthalmic compositions) is known in the art and includes, for example, vials (e.g., sealed vials), jars, ampoules, bottles, jars, flexible packaging (e.g., sealed Mylar or plastic bags), etc.These articles of manufacture may be further sterilized and/or sealed.

Additionally, kits are provided that contain the pharmaceutical compositions (or articles of manufacture) described herein, which may further include instructions regarding how to use the compositions, such as for the applications described herein. The kits described herein may also include other materials desirable from a commercial and user standpoint, including other buffers, diluents, filters, needles, syringes, and package inserts containing instructions for practicing any of the methods described herein.

VII. Therapeutic Uses
The pharmaceutical compositions described herein (including pharmaceutical compositions comprising variant CD80 polypeptides, immunomodulatory proteins, conjugates, and modified cells described herein) can be used in a wide variety of therapeutic applications, such as treating disease. For example, in some embodiments, the pharmaceutical compositions are used to treat inflammatory or autoimmune disorders, cancer, organ transplants, viral infections, and/or bacterial infections in mammals. The pharmaceutical compositions can modulate (e.g., enhance or reduce) immune responses to treat disease. In some embodiments, the methods provided are applicable to therapeutic administration of variant CD80 polypeptides, immunomodulatory proteins, conjugates, modified cells, and infectious agents described herein. Given the disclosure provided, it is within the level of ordinary skill in the art to select a format for an indication depending on the type of immune response modulation (e.g., enhance or reduce) desired.

In some embodiments, pharmaceutical compositions provided herein are administered that stimulate an immune response, which can be useful, for example, in the treatment of cancer, viral infection, or bacterial infection. In some embodiments, the pharmaceutical composition contains a variant CD80 polypeptide in a format that exhibits agonist activity of its cognate binding partner CD28 and/or stimulates or initiates costimulatory signaling through CD28. Exemplary formats of CD80 polypeptides for use in connection with such therapeutic applications include, for example, immunomodulatory proteins or "stacks" of variant CD80 polypeptides and IgSF domains or variants thereof (also referred to as "tumor-localized IgSF domains") that bind to tumor antigens (e.g., Nkp30 or affinity-modified variants), conjugates containing variant CD80 polypeptides linked to tumor targeting moieties (also referred to as tumor-localizing moieties), modified cells that express a transmembrane immunomodulatory protein, or infectious agents that include a nucleic acid molecule encoding a transmembrane immunomodulatory protein, for example, for expression in infected cells (e.g., tumor cells or APCs, e.g., dendritic cells) of the transmembrane immunomodulatory protein.

In some embodiments, the pharmaceutical compositions can be used to inhibit the growth of mammalian cancer cells (e.g., human cancer cells). A method of treating cancer can include administering to a subject having cancer an effective amount of any of the pharmaceutical compositions described herein. An effective amount of the pharmaceutical composition can be administered to inhibit, stop, or reverse the progression of the cancer.

Cancers that can be treated by the pharmaceutical compositions and methods of treatment described herein include, but are not limited to, melanoma, bladder cancer, hematological malignancies (leukemia, lymphoma, myeloma), liver cancer, brain cancer, kidney cancer, breast cancer, pancreatic cancer (adenocarcinoma), colon cancer, lung cancer (small cell lung cancer and non-small cell lung cancer), spleen cancer, thymus or blood cell cancer (i.e., leukemia), prostate cancer, testicular cancer, ovarian cancer, uterine cancer, gastric cancer, musculoskeletal cancer, head and neck cancer, gastrointestinal cancer, germ cell cancer, or endocrine and neuroendocrine cancer. In some embodiments, the cancer is Ewing's sarcoma. In some embodiments, the cancer is selected from melanoma, lung cancer, bladder cancer, and hematological malignancies. In some embodiments, the cancer is lymphoma, lymphocytic leukemia, myeloid leukemia, cervical cancer, neuroblastoma, or multiple myeloma.

Human cancer cells can be treated in vivo or ex vivo. In ex vivo treatment of a human patient, tissue or bodily fluids containing cancer cells are treated outside the body, and then the tissue or fluids are reintroduced back into the patient. In some embodiments, cancer is treated in vivo in a human patient by administration of a therapeutic composition to the patient.

In some embodiments, the pharmaceutical composition is administered as a monotherapy (i.e., as a single agent) or as a combination therapy (i.e., in combination with one or more additional anti-cancer agents, e.g., chemotherapeutic agents, cancer vaccines, or immune checkpoint inhibitors). In some embodiments, the pharmaceutical composition can also be administered with radiation therapy.

In some embodiments, the pharmaceutical composition suppresses an immune response, which can be useful in the treatment of inflammatory or autoimmune disorders, or organ transplantation. In some embodiments, the pharmaceutical composition contains a variant CD80 polypeptide in a format that exhibits antagonist activity of its cognate binding partner CD28 and/or blocks or inhibits costimulatory signaling through CD28. Exemplary formats of CD80 polypeptides for use in connection with such therapeutic applications include, for example, a variant CD80 polypeptide that is soluble (e.g., a variant CD80-Fc fusion protein), an immunomodulatory protein or "stack" of a variant CD80 polypeptide and another IgSF domain, including soluble forms thereof that are Fc fusions, modified cells that express a secreted immunomodulatory protein, or an infectious agent that includes a nucleic acid molecule encoding a secreted immunomodulatory protein, for expression and secretion in infected cells (e.g., tumor cells or APCs, e.g., dendritic cells) of the secreted immunomodulatory protein, for example.

In some embodiments, the inflammatory or autoimmune disorder is antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis, vasculitis, an autoimmune skin disease, transplant surgery, a rheumatic disease, an inflammatory gastrointestinal disease, an inflammatory eye disease, an inflammatory neurological disease, an inflammatory lung disease, an inflammatory endocrine disease, or an autoimmune blood disease.

In some embodiments, the inflammatory and autoimmune disorders that can be treated by the pharmaceutical compositions described herein include Addison's disease, allergies, alopecia areata, Alzheimer's disease, antineutrophil cytoplasmic antibody (ANCA) associated vasculitis, ankylosing spondylitis, antiphospholipid syndrome (Hughes syndrome), asthma, atherosclerosis, rheumatoid arthritis, autoimmune hemolytic anemia, autoimmune hepatitis, autoimmune inner ear disease, autoimmune lymphoproliferative syndrome, autoimmune myocarditis, autoimmune oophoritis, autoimmune orchitis, azoospermia, Behcet's disease, Berger's disease, bullous pemphigoid, cardiomyopathy, cardiovascular disease, celiac sprue/celiac disease, chronic fatigue immune dysregulation syndrome (CFIDS), chronic idiopathic polyneuropathy, chronic inflammatory demyelinating polyradiculoneuropathy (CIPD), chronic relapsing ... Guillain-Barré syndrome, Churg-Strauss syndrome (CSS), cicatricial pemphigoid, cold agglutinin disease (CAD), chronic obstructive pulmonary disease (COPD), CREST syndrome, Crohn's disease, dermatitis, dermatitis herpetiformis, dermatomyositis, diabetes, discoid lupus, eczema, epidermolysis bullosa acquisita, essential mixed cryoglobulinemia, Evans syndrome, exophthalmos, fibromyalgia, Goodpasture's syndrome, Graves' disease, Hashimoto's thyroiditis, idiopathic pulmonary fibrosis, idiopathic thrombocytopenic purpura (ITP), IgA nephropathy, immunoproliferative diseases or disorders, inflammatory bowel disease (IBD), interstitial lung disease, juvenile arthritis, juvenile idiopathic arthritis (JIA), Kawasaki disease, Lambert-Eaton myasthenic syndrome, lichen planus, lupus nephritis, lymphocytic hypophysitis lypophisitis), Meniere's disease, Miller Fisher syndrome/acute disseminated encephalomyeloradiculopathy, mixed connective tissue disease, multiple sclerosis (MS), muscular rheumatism, myalgic encephalomyelitis (ME), myasthenia gravis, eye inflammation, pemphigus foliaceus, pemphigus vulgaris, pernicious anaemia, polyarteritis nodosa, polychondritis, polyglandular syndrome syndromes) (Whittaker syndrome), polymyalgia rheumatica, polymyositis, primary agammaglobulinemia, primary biliary cirrhosis/autoimmune cholangiopathy, psoriasis, psoriatic arthritis, Raynaud's phenomenon, Reiter's syndrome/reactive joints, restenosis, rheumatic fever, rheumatic disease, sarcoidosis, Schmidt's syndrome, scleroderma, Sjogren's syndrome, stiff-man syndrome, systemic lupus erythematosus (SLE), systemic scleroderma, Takayasu's arteritis, temporal arteritis/giant cell arteritis, thyroiditis, type 1 diabetes, ulcerative colitis, uveitis, vasculitis, vitiligo, interstitial bowel disease, or Wegener's granulomatosis. In some embodiments, the inflammatory or autoimmune disorder is selected from interstitial bowel disease, transplantation, Crohn's disease, ulcerative colitis, multiple sclerosis, asthma, rheumatoid arthritis, and psoriasis.

In some embodiments, the pharmaceutical compositions are administered to modulate an autoimmune pathology. For example, suppressing an immune response can be beneficial in methods for inhibiting the rejection of a tissue, cell, or organ transplant from a donor by a recipient. Thus, in some embodiments, the pharmaceutical compositions described herein are used to limit or prevent transplant-associated or transplant-related diseases or disorders, such as graft-versus-host disease (GVHD). In some embodiments, the pharmaceutical compositions are used to suppress autoimmune rejection of transplanted or grafted bone marrow, organs, skin, muscle, neurons, pancreatic islets, or parenchymal cells.

Pharmaceutical compositions and methods comprising the modified cells described herein can be used in adoptive cell transfer applications. In some embodiments, cell compositions comprising the modified cells can be used in related methods, for example, to modulate immune activity, for example, in immunotherapeutic approaches to the treatment of mammalian cancer, or in other embodiments, autoimmune disorders. The methods employed generally include contacting the TIP of the present invention with a mammalian cell under conditions permissive for specific binding of the affinity-modified IgSF domain and modulation of immune activity of the mammalian cell. In some embodiments, immune cells (e.g., tumor-infiltrating lymphocytes (TILs), T cells (including CD8 ⁺ or CD4 ⁺ T cells), or APCs) are modified to express as membrane proteins and/or soluble variant CD80 polypeptides, immune-modulating proteins, or conjugates as described herein. The modified cells can then be contacted with mammalian cells, such as APCs, second lymphocytes, or tumor cells, whose immune activity is desired to be modulated, under conditions permissive for specific binding of the affinity-modified IgSF domain to counter structures on the mammalian cell that can modulate immune activity in the mammalian cell. The cells can be contacted in vivo or ex vivo.

In some embodiments, the modified cells are autologous cells. In other embodiments, the cells are allogeneic cells. In some embodiments, the cells are autologous modified cells that are reinfused into the mammal from which they were isolated. In some embodiments, the cells are allogeneic modified cells that are infused into the mammal. In some embodiments, the cells are harvested from the patient's blood or tumor, modified to express a polypeptide (e.g., a variant CD80 polypeptide, an immunomodulatory protein, or a conjugate as described herein), expanded in an in vitro culture system (e.g., by stimulating the cells), and reinfused into the patient to mediate tumor destruction. In some embodiments, the method is carried out by adoptive cell transfer, in which cells expressing TIP (e.g., T cells) are infused back into the patient. In some embodiments, the therapeutic compositions and methods of the invention are used in the treatment of a mammalian patient with cancer, such as lymphoma, lymphocytic leukemia, myeloid leukemia, cervical cancer, neuroblastoma, or multiple myeloma.

またはその保存的アミノ酸置換である、態様1～11のいずれかのバリアントCD80ポリペプチド。
13. 前記1つまたは複数のアミノ酸置換が、

である、態様1～12のいずれかのバリアントCD80ポリペプチド。
14. SEQ ID NO:55～108、280～346、414～475のうちのいずれかもしくはその特異的結合断片に示されるアミノ酸配列、またはSEQ ID NO:55～108、280～346、414～475のうちのいずれかもしくはその特異的結合断片に対して少なくとも95％の配列同一性を示しかつ前記1つまたは複数のアミノ酸置換を含有するアミノ酸配列を含む、態様1～13のいずれかのバリアントCD80ポリペプチド。
15. 前記IgVドメインまたはその特異的結合断片を含む、態様1～14のいずれかのバリアントCD80ポリペプチド。
16. 前記IgVドメインまたはその特異的断片が、前記バリアントCD80ポリペプチドの唯一のCD80部分である、態様1～15のいずれかのバリアントCD80ポリペプチド。
17. 前記IgCドメインまたはその特異的断片が、前記バリアントCD80ポリペプチドの唯一のCD80部分である、態様1～14のいずれかのバリアントCD80ポリペプチド。
18. SEQ ID NO:153～195、347、373～386、476～477のいずれかもしくはその特異的結合断片に示されるアミノ酸配列、SEQ ID NO:153～195、347、373～386、476～477のいずれかもしくはその特異的結合断片に対して少なくとも95％の配列同一性を示しかつ前記1つまたは複数のアミノ酸置換を含有するアミノ酸配列を含む、態様1～17のいずれかのバリアントCD80ポリペプチド。
19. CD28、PD-L1、またはCTLA-4のエクトドメインに、前記未改変CD80ポリペプチドと比較して向上した親和性で特異的に結合する、態様1～17のいずれかのバリアントCD80ポリペプチド。
20. CD28、PD-L1、またはCTLA-4のエクトドメインに、該エクトドメインに対する前記未改変CD80の結合性と比較して向上した選択性で特異的に結合する、態様1～19のいずれかのバリアントCD80ポリペプチド。
21. 前記選択性の向上が、CD28、PD-L1、およびCTLA-4の中から選択される1つの同族結合パートナーに対する結合性と、該同族結合パートナーのうちの別のものに対する結合性との比率が、該1つの同族結合パートナーに対する前記未改変CD80ポリペプチドの結合性と、該同族結合パートナーのうちの該別のものに対する該未改変CD80ポリペプチドの結合性との比率と比較して大きいことを含む、態様20のバリアントCD80ポリペプチド。
22. CD28のエクトドメインに、該CD28のエクトドメインに対する前記未改変CD80の結合性と比較して向上した選択性で特異的に結合する、態様19または態様20のバリアントCD80ポリペプチド。
23. 前記選択性の向上が、CD28に対する結合性と、PD-L1またはCTLA-4に対する結合性との比率が、CD28に対する前記未改変CD80ポリペプチドの結合性と、PD-L1またはCTLA-4に対する該未改変CD80ポリペプチドの結合性との比率と比較して大きいことを含む、態様22のバリアントCD80ポリペプチド。
24. 前記比率が、少なくともまたは少なくとも約1.5倍、2.0倍、3.0倍、4.0倍、5倍、10倍、15倍、20倍、30倍、40倍、50倍、またはそれ以上大きい、態様21または態様23のバリアントCD80ポリペプチド。
25. CD28のエクトドメインに、前記未改変CD80ポリペプチドと比較して向上した親和性で特異的に結合する、態様1～24のいずれかのバリアントCD80ポリペプチド。
26. 前記エクトドメインに対する向上した親和性が、前記エクトドメインに対する前記未改変CD80の結合親和性と比較して、1.2倍、1.5倍、2倍、3倍、4倍、5倍、6倍、7倍、8倍、9倍、10倍、20倍、30倍、40倍、50倍、または60倍より大きく向上している、態様19または態様25のバリアントCD80ポリペプチド。
27. 前記1つまたは複数のアミノ酸置換が、SEQ ID NO:28のナンバリングを基準にした位置12、18、20、29、31、36、40、41、43、52、59、60、63、67、70、77、81、87、88、89、90、91、92、93、107、109、114、117、118、120、122、127、130、144、169、177、または199に対応する、態様1～26のいずれかのバリアントCD80ポリペプチド。
28. 前記1つまたは複数のアミノ酸置換が、

ならびにその保存的アミノ酸置換からなる群より選択される、態様1～27のいずれかのバリアントCD80ポリペプチド。
29. 前記1つまたは複数のアミノ酸置換が、

である、態様1～28のいずれかのバリアントCD80ポリペプチド。
30. PD-L1のエクトドメインに、前記未改変CD80ポリペプチドと比較して向上した親和性で特異的に結合する、態様1～29のいずれかのバリアントCD80ポリペプチド。
31. 前記エクトドメインに対する向上した親和性が、前記エクトドメインに対する前記未改変CD80の結合親和性と比較して、1.2倍、1.5倍、2倍、3倍、4倍、5倍、6倍、7倍、8倍、9倍、10倍、20倍、30倍、40倍、50倍、または60倍より大きく向上している、態様30のバリアントCD80ポリペプチド。
32. PD-L1のエクトドメインに、該エクトドメインに対する前記未改変CD80の結合性と比較して向上した選択性で特異的に結合する、態様1～31のいずれかのバリアントCD80ポリペプチド。
33. 前記選択性の向上が、PD-L1に対する結合性と、CD28またはCTLA-4に対する結合性との比率が、PD-L1に対する前記未改変CD80ポリペプチドの結合性と、CD28またはCTLA-4に対する該未改変CD80ポリペプチドの結合性との比率と比較して大きいことを含む、態様32のバリアントCD80ポリペプチド。
34. 前記比率が、少なくともまたは少なくとも約1.5倍、2.0倍、3.0倍、4.0倍、5倍、10倍、15倍、20倍、30倍、40倍、50倍、またはそれ以上大きい、態様33のバリアントCD80ポリペプチド。
35. 前記1つまたは複数のアミノ酸置換が、SEQ ID NO:28のナンバリングを基準にした位置12、18、29、31、33、36、38、40、41、42、43、47、48、59、64、67、68、70、77、81、85、87、88、89、90、91、92、93、94、97、104、109、115、117、118、120、122、126、130、133、140、144、148、149、168、178、183、188または193に対応する、態様1～34のいずれかのバリアントCD80ポリペプチド。
36. 前記1つまたは複数のアミノ酸置換が、

ならびにその保存的アミノ酸置換からなる群より選択される、態様1～35のいずれかのバリアントCD80ポリペプチド。
37. 前記1つまたは複数のアミノ酸置換が、

である、態様1～36のいずれかのバリアントCD80ポリペプチド。
38. CD28のエクトドメインおよびPD-L1のエクトドメインに、前記未改変CD80ポリペプチドと比較して向上した親和性で特異的に結合する、態様1～37のいずれかのバリアントCD80ポリペプチド。
39. 前記1つまたは複数のアミノ酸置換が、SEQ ID NO:28のナンバリングを基準にした位置12、18、36、40、43、59、77、88、89、90、91、92、93、94、109、118、122、または177に対応する、態様1～38のいずれかのバリアントCD80ポリペプチド。
40. 前記1つまたは複数のアミノ酸置換が、

およびその保存的アミノ酸置換からなる群より選択される、態様1～39のいずれかのバリアントCD80ポリペプチド。
41. 前記1つまたは複数のアミノ酸置換が、

である、態様1～40のいずれかのバリアントCD80ポリペプチド。
42. CTLA-4のエクトドメインに、前記未改変CD80ポリペプチドと比較して向上した親和性で特異的に結合する、態様1～41のいずれかのバリアントCD80ポリペプチド。
43. 前記エクトドメインに対する向上した親和性が、前記エクトドメインに対する前記未改変CD80の結合親和性と比較して、1.2倍、1.5倍、2倍、3倍、4倍、5倍、6倍、7倍、8倍、9倍、10倍、20倍、30倍、40倍、50倍、または60倍より大きく向上している、態様42のバリアントCD80ポリペプチド。
44. CTLA-4のエクトドメインに、該エクトドメインに対する前記未改変CD80の結合性と比較して向上した選択性で特異的に結合する、態様1～43のいずれかのバリアントCD80ポリペプチド。
45. 選択性の向上が、CTLA-4に対する結合性と、CD28またはPD-L1に対する結合性との比率が、CTLA-4に対する前記未改変CD80ポリペプチドの結合性と、CD28またはPD-L1に対する該未改変CD80ポリペプチドの結合性との比率と比較して大きいことを含む、態様44のバリアントCD80ポリペプチド。
46. 前記比率が、少なくともまたは少なくとも約1.5倍、2.0倍、3.0倍、4.0倍、5倍、10倍、15倍、20倍、30倍、40倍、50倍またはそれ以上大きい、態様45のバリアントCD80ポリペプチド。
47. 前記1つまたは複数のアミノ酸置換が、SEQ ID NO:28のナンバリングを基準にした位置4、29、31、36、40、41、52、67、68、70、87、88、89、90、91、92、93、107、109、110、118、120、130、144、または169に対応する、態様1～46のいずれかのバリアントCD80ポリペプチド。
48. 前記1つまたは複数のアミノ酸置換が、

ならびにその保存的アミノ酸置換からなる群より選択される、態様1～47のいずれかのバリアントCD80ポリペプチド。
49. 前記1つまたは複数のアミノ酸置換が、

である、態様1～48のいずれかのバリアントCD80ポリペプチド。
50. CD28のエクトドメインおよびCTLA-4のエクトドメインに、前記未改変CD80ポリペプチドと比較して向上した親和性で特異的に結合する、態様1～49のいずれかのバリアントCD80ポリペプチド。
51. 前記1つまたは複数のアミノ酸置換が、SEQ ID NO:28の位置36、40、52、70、88、89、90、91、92、93、107、118、120、130、144、または169に対応する、態様1～50のいずれかのバリアントCD80ポリペプチド。
52. 前記1つまたは複数のアミノ酸置換が、

ならびにその保存的アミノ酸置換からなる群より選択される、態様1～51のいずれかのバリアントCD80ポリペプチド。
53. 前記1つまたは複数のアミノ酸置換が、

である、態様1～52のいずれかのバリアントCD80ポリペプチド。
54. PD-L1のエクトドメインおよびCTLA-4のエクトドメインに、前記未改変CD80ポリペプチドと比較して向上した親和性で特異的に結合する、態様1～53のいずれかのバリアントCD80ポリペプチド。
55. 前記1つまたは複数のアミノ酸置換が、SEQ ID NO:28の位置29、31、36、40、41、67、70、87、88、89、90、91、92、93、109、118、120、122、130、または178に対応する、態様1～54のいずれかのバリアントCD80ポリペプチド。
56. 前記1つまたは複数のアミノ酸置換が、

およびその保存的アミノ酸置換からなる群より選択される、態様1～55のいずれかのバリアントCD80ポリペプチド。
57. 前記1つまたは複数のアミノ酸置換が、

である、態様1～56のいずれかのバリアントCD80ポリペプチド。
58. CD28のエクトドメイン、PD-L1のエクトドメイン、およびCTLA-4のエクトドメインに、前記未改変CD80ポリペプチドと比較して向上した親和性で特異的に結合する、態様1～57のいずれかのバリアントCD80ポリペプチド。
59. 前記1つまたは複数のアミノ酸置換が、SEQ ID NO:28の位置36、40、88、89、90、91、92、または93に対応する、態様1～58のいずれかのバリアントCD80ポリペプチド。
60. 前記1つまたは複数のアミノ酸置換が、K36G、L40M、E88D、K89R、D90K、A91G、F92Y、K93R、およびその保存的アミノ酸置換からなる群より選択される、態様1～59のいずれかのバリアントCD80ポリペプチド。
61. 前記1つまたは複数のアミノ酸置換が、

である、態様1～60のいずれかのバリアントCD80ポリペプチド。
62. CD28のエクトドメインまたはPD-L1のエクトドメインに、前記未改変CD80ポリペプチドと比較して向上した親和性で特異的に結合し、かつ、
CTLA-4のエクトドメインに、前記未改変CD80ポリペプチドと比較して低下した親和性で特異的に結合する、
態様1～41のいずれかのバリアントCD80ポリペプチド。
63. 前記1つまたは複数のアミノ酸置換が、SEQ ID NO:28のナンバリングを基準にした位置29、31、33、36、38、41、42、43、47、63、67、70、81、85、87、88、89、90、91、92、93、94、109、114、118、120、127、130、144、148、または149に対応する、態様1～41および62のいずれかのバリアントCD80ポリペプチド。
64. 前記1つまたは複数のアミノ酸置換が、

ならびにその保存的アミノ酸置換からなる群より選択される、態様1～41、62、および63のいずれかのバリアントCD80ポリペプチド。
65. 前記1つまたは複数のアミノ酸置換が、

である、態様1～41および62～64のいずれかのバリアントCD80ポリペプチド。
66. CD28のエクトドメインに、前記未改変CD80ポリペプチドと比較して向上した親和性で特異的に結合し、かつ、
CTLA-4のエクトドメインに、前記未改変CD80ポリペプチドと比較して低下した親和性で特異的に結合する、
態様1～41および62～65のいずれかのバリアントCD80ポリペプチド。
67. 前記1つまたは複数のアミノ酸置換が、SEQ ID NO:28の位置63、70、81、87、91、114、118、120、127、または130に対応する、態様1～41および62～66のいずれかのバリアントCD80ポリペプチド。
68. 前記1つまたは複数のアミノ酸置換が、

ならびにその保存的アミノ酸置換からなる群より選択される、態様1～41、および62～67のいずれかのバリアントCD80ポリペプチド。
69. 前記1つまたは複数のアミノ酸置換が、

である、態様1～41、および62～67のいずれかのバリアントCD80ポリペプチド。
70. PD-L1のエクトドメインに、前記未改変CD80ポリペプチドと比較して向上した親和性で特異的に結合し、かつ、
CTLA-4のエクトドメインに、前記未改変CD80ポリペプチドと比較して低下した親和性で特異的に結合する、
態様1～41および62～69のいずれかのバリアントCD80ポリペプチド。
71. 前記1つまたは複数のアミノ酸置換が、SEQ ID NO:28の位置29、31、33、36、38、41、42、43、47、67、70、81、85、87、88、89、90、91、92、93、94、109、118、120、144、148、または149に対応する、態様1～41および62～70のいずれかのバリアントCD80ポリペプチド。
72. 前記1つまたは複数のアミノ酸置換が、

ならびにその保存的アミノ酸置換からなる群より選択される、態様1～41、および62～71のいずれかのバリアントCD80ポリペプチド。
73. 前記1つまたは複数のアミノ酸置換が、

である、態様1～41、および62～72のいずれかのバリアントCD80ポリペプチド。
74. CD28のエクトドメインおよびPD-L1のエクトドメインに、前記未改変CD80ポリペプチドと比較して向上した親和性で特異的に結合し、かつ、
CTLA-4のエクトドメインに、前記未改変CD80ポリペプチドと比較して低下した親和性で特異的に結合する、
態様1～41および62～73のいずれかのバリアントCD80ポリペプチド。
75. 前記1つまたは複数のアミノ酸置換が、SEQ ID NO:28の位置70、81、87、91、または120に対応する、態様1～41および62～74のいずれかのバリアントCD80ポリペプチド。
76. 前記1つまたは複数のアミノ酸置換が、L70Q、Y87N、A91G、およびT120S、ならびにその保存的アミノ酸置換からなる群より選択される、態様1～41、および62～75のいずれかのバリアントCD80ポリペプチド。
77. CD28がヒトCD28である、態様10～76のいずれかのバリアントCD80ポリペプチド。
78. PD-L1がヒトPD-L1である、態様10～77のいずれかのバリアントCD80ポリペプチド。
79. CTLA-4がヒトCTLA-4である、態様10～78のいずれかのバリアントCD80ポリペプチド。
80. 前記結合活性が、前記未改変CD80ポリペプチドと比較して、1.2倍、1.5倍、2倍、3倍、4倍、5倍、6倍、7倍、8倍、9倍、10倍、20倍、30倍、40倍、または50倍より大きく変化（向上または低下）している、態様1～79のいずれかのバリアントCD80ポリペプチド。
81. 可溶性タンパク質である、態様1～80のいずれかのバリアントCD80ポリペプチド。
82. 多量体化ドメインに連結されている、態様1～81のいずれかのバリアントCD80ポリペプチド。
83. 多量体化ドメインに連結された第一のバリアントCD80ポリペプチドおよび多量体化ドメインに連結された第二のバリアントCD80ポリペプチドを含む、多量体ポリペプチド、任意で二量体ポリペプチドである、態様1～81のいずれかのバリアントCD80ポリペプチド。
84. 前記第一のバリアントCD80ポリペプチドおよび前記第二のバリアントCD80ポリペプチドが、同じであるかまたは異なる、態様83のバリアントCD80ポリペプチド。
85. 前記多量体化ドメインが、エフェクター機能が低下しているFcドメインまたはそのバリアントである、態様82～84のいずれかのバリアントCD80ポリペプチド。
86. 前記ポリペプチドの生物学的半減期を延長させる部分に連結されている、態様1～85のいずれかのバリアントCD80ポリペプチド。
87. エフェクター機能が低下しているFcドメインまたはそのバリアントに連結されている、態様1～86のいずれかのバリアントCD80ポリペプチド。
88. 前記Fcドメインが、哺乳動物のもの、任意でヒトのものであるか；または
哺乳動物のもの、任意でヒトのものである未改変Fcドメインと比較して、前記バリアントFcドメインが1つまたは複数のアミノ酸改変を含む、
態様85～87のいずれかのバリアントCD80ポリペプチド。
89. 前記Fcドメインまたはそのバリアントが、SEQ ID NO:226もしくはSEQ ID NO:227に示されるアミノ酸配列、またはSEQ ID NO:226もしくはSEQ ID NO:227に対して少なくとも85％の配列同一性を示すアミノ酸配列を含む、態様85～88のいずれかのバリアントCD80ポリペプチド。
90. リンカーを介して間接的に連結されている、態様82～89のいずれかのバリアントCD80ポリペプチド。
91. 前記バリアントCD80ポリペプチドの細胞外ドメイン（ECD）またはその特異的結合断片に連結された膜貫通ドメインをさらに含む膜貫通型免疫調節タンパク質である、態様1～90のいずれかのバリアントCD80ポリペプチド。
92. 前記膜貫通ドメインが、SEQ ID NO:1の243～263番目の残基として示されるアミノ酸配列、またはSEQ ID NO:1の243～263番目の残基に対して少なくとも85％の配列同一性を示すその機能的バリアントを含む、態様91のバリアントCD80ポリペプチド。
93. 前記膜貫通ドメインに連結された細胞質ドメインをさらに含む、態様91または態様92のバリアントCD80ポリペプチド。
94. 前記細胞質ドメインが、SEQ ID NO:1の264～288番目の残基として示されるアミノ酸配列、または、SEQ ID NO:1の264～288番目の残基に対して少なくとも85％の配列同一性を示すその機能的バリアントを含む、態様93のバリアントCD80ポリペプチド。
95. インビトロ初代T細胞アッセイにおいて、前記未改変CD80と比べてIFN-γ（インターフェロン-γ）発現を増加させる、態様1～94のいずれかのバリアントCD80ポリペプチド。
96. インビトロ初代T細胞アッセイにおいて、前記未改変CD80と比べてIFN-γ（インターフェロン-γ）発現を減少させる、態様1～95のいずれかのバリアントCD80ポリペプチド。
97. 脱グリコシル化されている、態様1～96のいずれかのバリアントCD80ポリペプチド。
98. 免疫グロブリンスーパーファミリー（IgSF）ドメインを含む第二のポリペプチドに連結された態様1～97のいずれかのバリアントCD80を含む、免疫調節ポリペプチド。
99. 前記IgSFドメインが、未改変または野生型IgSFドメインと比較して、親和性が改変されており、かつその同族結合パートナーのうちの1つまたは複数に対する変化した結合性を示す、態様98の免疫調節ポリペプチド。
100. 前記IgSFドメインが、未改変または野生型IgSFドメインと比較して、その同族結合パートナーのうちの1つまたは複数に対する向上した結合性を示す、態様99の免疫調節ポリペプチド。
101. 前記バリアントCD80が第一のバリアントCD80であり、前記第二のポリペプチドのIgSFドメインが態様1～97のいずれかの第二のバリアントCD80由来のIgSFドメインであり、該第一および該第二のバリアントCD80が、同じであるかまたは異なる、態様98～100のいずれかの免疫調節ポリペプチド。
102. 前記バリアントCD80ポリペプチドが、CD28、PD-L1、またはCTLA-4に特異的に結合することができ、前記第二のポリペプチドのIgSFドメインが、該バリアントCD80ポリペプチドが特異的に結合するもの以外の同族結合パートナーに結合することができる、態様98～101のいずれかの免疫調節ポリペプチド。
103. 前記バリアントCD80ポリペプチドが、CD28またはPD-L1に特異的に結合することができ、前記IgSFドメインが、該バリアントCD80ポリペプチドが特異的に結合するもの以外の同族結合パートナーに結合することができる、態様98～102のいずれかの免疫調節ポリペプチド。
104. 前記バリアントCD80ポリペプチドが、CD28またはCTLA-4に特異的に結合することができ、前記IgSFドメインが、該バリアントCD80ポリペプチドが特異的に結合するもの以外の同族結合パートナーに結合することができる、態様98～103のいずれかの免疫調節ポリペプチド。
105. 前記バリアントCD80ポリペプチドが、CTLA-4またはPD-L1に特異的に結合することができ、前記IgSFドメインが、該バリアントCD80ポリペプチドが特異的に結合するもの以外の同族結合パートナーに結合することができる、態様98～103のいずれかの免疫調節タンパク質。
106. 前記IgSFドメインが、B7ファミリーのメンバー由来である、態様98～105のいずれかの免疫調節ポリペプチド。
107. 前記IgSFドメインが、腫瘍上に発現するリガンドに結合する腫瘍局在化部分である、態様82～90のいずれかの免疫調節ポリペプチド。
108. 前記リガンドがB7H6である、態様107の免疫調節ポリペプチド。
109. 前記IgSFドメインがNKp30由来である、態様107または態様108の免疫調節ポリペプチド。
110. 前記IgSFドメインが、未改変または野生型IgSFドメインと比較して、親和性が改変されており、かつその同族結合パートナーのうちの1つまたは複数に対する向上した結合性を示す、態様98～109のいずれかの免疫調節ポリペプチド。
111. 前記IgSFドメインが、IgVドメインであるかまたはIgVドメインを含む、態様98～110のいずれかの免疫調節ポリペプチド。
112. 前記バリアントCD80ポリペプチドが、IgVドメインであるかまたはIgVドメインを含む、態様98～111のいずれかの免疫調節ポリペプチド。
113. 前記バリアントCD80ポリペプチドおよび前記IgSFドメインを含む第二のポリペプチドのうちの一方または両方に連結された多量体化ドメインを含む、態様98～112のいずれかの免疫調節ポリペプチド。
114. 前記多量体化ドメインが、エフェクター機能が低下しているFcドメインまたはそのバリアントである、態様113の免疫調節タンパク質。
115. 二量体である、態様98～114のいずれかの免疫調節ポリペプチド。
116. ホモ二量体である、態様115の免疫調節タンパク質。
117. ヘテロ二量体である、態様115の免疫調節タンパク質。
118. ある部分に連結された、態様1～98のいずれかのバリアントCD80または態様99～117のいずれかの免疫調節ポリペプチドを含む、コンジュゲート。
119. 前記部分が、細胞の表面上の分子に特異的に結合する標的化部分である、態様118のコンジュゲート。
120. 前記標的化部分が、免疫細胞の表面上の分子に特異的に結合する、態様119のコンジュゲート。
121. 前記免疫細胞が、抗原提示細胞またはリンパ球である、態様120のコンジュゲート。
122. 前記標的化部分が、腫瘍の表面上の分子に結合する腫瘍局在化部分である、態様119のコンジュゲート。
123. 前記部分が、タンパク質、ペプチド、核酸、低分子、またはナノ粒子である、態様118～122のいずれかのコンジュゲート。
124. 前記部分が、抗体または抗原結合断片である、態様118～123のいずれかのコンジュゲート。
125. 二価、四価、六価、または八価である、態様118～124のいずれかのコンジュゲート。
126. 態様1～98のいずれかのバリアントCD80ポリペプチドまたは態様99～117のいずれかの免疫調節ポリペプチドをコードする、核酸分子。
127. 合成核酸である、態様126の核酸分子。
128. cDNAである、態様126または態様127の核酸分子。
129. 態様125～128のいずれかの核酸分子を含む、ベクター。
130. 発現ベクターである、態様129のベクター。
131. 哺乳動物ベクターまたはウイルスベクターである、態様129または態様130のベクター。
132. 態様129～131のいずれかのベクターを含む、細胞。
133. 哺乳動物細胞である、態様132の細胞。
134. ヒト細胞である、態様133の細胞。
135. バリアントCD80ポリペプチドまたは免疫調節タンパク質を生産する方法であって、態様125～128のいずれかの核酸分子または態様129～131のいずれかのベクターを、宿主細胞において該タンパク質が発現する条件下で宿主細胞に導入する工程を含む、方法。
136. 前記バリアントCD80ポリペプチドまたは前記免疫調節タンパク質を前記細胞から単離または精製する工程をさらに含む、態様135の方法。
137. バリアントCD80ポリペプチドを発現する細胞を改変する方法であって、態様1～98のいずれかのバリアントCD80ポリペプチドをコードする核酸分子を、宿主細胞において該ポリペプチドが発現する条件下で宿主細胞に導入する工程を含む、方法。
138. 態様1～98のいずれかのバリアントCD80ポリペプチド、態様99～117のいずれかの免疫調節タンパク質、態様125～128のいずれかの核酸分子、または態様129～131のいずれかのベクターを発現する、改変細胞。
139. 前記バリアントCD80ポリペプチドまたは免疫調節ポリペプチドが、シグナルペプチドを含む、態様138の改変細胞。
140. 前記バリアントCD80ポリペプチドまたは免疫調節ポリペプチドが、膜貫通ドメインを含まず、かつ/または、前記細胞の表面上には発現しない、態様138または態様139の改変細胞。
141. 前記バリアントCD80ポリペプチドまたは免疫調節ポリペプチドが、前記改変細胞から分泌される、態様138～140のいずれかの改変細胞。
142. 膜貫通ドメインを含みかつ/または態様91～97のいずれかの膜貫通型免疫調節タンパク質である、バリアントCD80ポリペプチド
を含む、態様138または態様139の改変細胞。
143. 前記バリアントCD80ポリペプチドが、前記細胞の表面上に発現する、態様138、態様139、または態様142の改変細胞。
144. 免疫細胞である、態様138～143のいずれかの改変細胞。
145. 前記免疫細胞が、抗原提示細胞（APC）またはリンパ球である、態様144の改変細胞。
146. 初代細胞である、態様138～145のいずれかの改変細胞。
147. 哺乳動物細胞である、態様138～146のいずれかの改変細胞。
148. ヒト細胞である、態様138～147のいずれかの改変細胞。
149. 前記リンパ球がT細胞である、態様145～148のいずれかの改変細胞。
150. 前記APCが人工APCである、態様145～149のいずれかの改変細胞。
151. キメラ抗原受容体（CAR）または改変されたT細胞受容体をさらに含む、態様138～150のいずれかの改変細胞。
152. 態様1～97のいずれかのバリアントCD80ポリペプチドまたは態様98～117のいずれかの免疫調節ポリペプチドをコードする核酸分子を含む、感染性物質。
153. 前記コードされているバリアントCD80ポリペプチドまたは免疫調節ポリペプチドが、膜貫通ドメインを含まず、かつ/または、それを発現する細胞の表面上には発現しない、態様152の感染性物質。
154. 前記コードされているバリアントCD80ポリペプチドまたは免疫調節ポリペプチドが、それを発現する細胞から分泌される、態様152または態様153の感染性物質。
155. 前記コードされているバリアントCD80ポリペプチドが、膜貫通ドメインを含む、態様152の感染性物質。
156. 前記コードされているバリアントCD80ポリペプチドが、それを発現する細胞の表面上に発現する、態様152または態様155の感染性物質。
157. 細菌またはウイルスである、態様152～156のいずれかの感染性物質。
158. 前記ウイルスが腫瘍溶解性ウイルスである、態様157の感染性物質。
159. 前記腫瘍溶解性ウイルスが、アデノウイルス、アデノ随伴ウイルス、ヘルペスウイルス、単純ヘルペスウイルス、水疱性口腔ウイルス（Vesticular Stomatic virus）、レオウイルス、ニューカッスル病ウイルス、パルボウイルス、麻疹ウイルス、水疱性口内炎ウイルス（vesticular stomatitis virus）（VSV）、コクサッキーウイルス、またはワクシニアウイルスである、態様158の感染性物質。
160. 前記ウイルスが、樹状細胞（DC）を特異的に標的化し、かつ/または、樹状細胞指向性である、態様159の感染性物質。
161. 前記ウイルスが、改変されたシンドビスウイルスエンベロープ産物で偽型化されたレンチウイルスベクターである、態様160の感染性物質。
162. 標的細胞の死をもたらすかまたは免疫応答を増強もしくは強化することができるさらなる遺伝子産物をコードする核酸分子をさらに含む、態様152～161のいずれかの感染性物質。
163. 前記さらなる遺伝子産物が、抗がん剤、抗転移剤、抗血管新生剤、免疫調節分子、免疫チェックポイント阻害剤、抗体、サイトカイン、成長因子、抗原、細胞傷害性遺伝子産物、アポトーシス促進性遺伝子産物、抗アポトーシス性遺伝子産物、細胞マトリックス分解性遺伝子、組織再生のための遺伝子、またはヒト体細胞の多能性へのリプログラミングのための遺伝子から選択される、態様162の感染性物質。
164. 態様1～97のいずれかのバリアントCD80ポリペプチド、態様98～117のいずれかの免疫調節タンパク質、態様118～125のいずれかのコンジュゲート、態様138～151のいずれかの改変細胞、または態様152～163のいずれかの感染性物質を含む、薬学的組成物。
165. 薬学的に許容し得る賦形剤を含む、態様164の薬学的組成物。
166. 滅菌されている、態様164または165の薬学的組成物。
167. 態様164～166のいずれかの薬学的組成物をバイアル中に含む、製造品。
168. 前記バイアルが密閉されている、態様167の製造品
169. 態様164～166のいずれかの薬学的組成物と使用説明書とを含む、キット。
170. 態様167および168の製造品と使用説明書とを含む、キット。
171. 対象における免疫応答を調節する方法であって、態様164～166のいずれかの薬学的組成物を該対象に投与する工程を含む、方法。
172. 対象における免疫応答を調節する方法であって、態様138～151のいずれかの改変細胞を投与する工程を含む、方法。
173. 前記改変細胞が、前記対象にとって自家である、態様172の方法。
174. 前記改変細胞が、前記対象にとって同種である、態様172の方法。
175. 前記免疫応答の調節が、前記対象における疾患または病態を治療する、態様171～174のいずれかの方法。
176. 免疫応答が増強される、態様171～175のいずれかの方法。
177. 腫瘍局在化部分に連結されたバリアントCD80ポリペプチドを含む免疫調節タンパク質またはコンジュゲートが前記対象に投与される、態様171、175、および176のいずれかの方法。
178. 前記腫瘍局在化部分が、腫瘍抗原を認識する結合分子であるかまたはそれを含む、態様177の方法。
179. 前記結合分子が、抗体もしくはその抗原結合断片を含むか、または、野生型IgSFドメインもしくはそのバリアントを含む、態様178の方法。
180. 態様107～117のいずれかの免疫調節タンパク質または態様118～125のいずれかのコンジュゲートを含む薬学的組成物が前記対象に投与される、態様171および175～179のいずれかの方法。
181. 膜貫通型免疫調節タンパク質であるバリアントCD80ポリペプチドを含む改変細胞が前記対象に投与され、かつ/または、態様138、139および142～151のいずれかの改変細胞が前記対象に投与される、態様172～176のいずれかの方法。
182. 膜貫通型免疫調節タンパク質であるバリアントCD80ポリペプチドをコードする感染性物質が、任意で該感染性物質が腫瘍細胞または免疫細胞に感染しかつ該膜貫通型免疫調節タンパク質が該感染細胞の表面上に発現する条件下で、前記対象に投与される、態様171、175および176のいずれかの方法。
183. 前記膜貫通型免疫調節タンパク質が、態様91～97のいずれかの膜貫通型免疫調節タンパク質である、態様181または態様182のいずれかの方法。
184. 前記疾患または病態が、腫瘍またはがんである、態様171～183のいずれかの方法。
185. 前記疾患または病態が、黒色腫、肺がん、膀胱がん、血液悪性腫瘍、肝臓がん、脳がん、腎臓がん、乳がん、膵臓がん、大腸がん、脾臓がん、前立腺がん、精巣がん、卵巣がん、子宮がん、胃がん、筋骨格がん、頭頚部がん、消化器がん、生殖細胞がん、または内分泌および神経内分泌がんから選択される、態様171～184のいずれかの方法。
186. 免疫応答が低下する、態様171～175のいずれかの方法。
187. 可溶性であるバリアントCD80ポリペプチドまたは免疫調節タンパク質が前記対象に投与される、態様171、175および186のいずれかの方法。
188. 前記可溶性ポリペプチドまたは免疫調節タンパク質が、Fc融合タンパク質である、態様187の方法。
189. 態様1～90および95～97のいずれかのバリアントCD80ポリペプチドまたは態様98～106および111～117のいずれかの免疫調節タンパク質を含む薬学的組成物が前記対象に投与される、態様171、175および186～188のいずれかの方法。
190. 分泌性バリアントCD80ポリペプチドを含む改変細胞が前記対象に投与される、態様172～175および186のいずれかの方法。
191. 態様138～141および144～151のいずれかの改変細胞が前記対象に投与される、態様172～175および186～188のいずれかの方法。
192. 分泌性免疫調節タンパク質であるバリアントCD80ポリペプチドをコードする感染性物質が、任意で該感染性物質が腫瘍細胞または免疫細胞に感染しかつ該分泌性免疫調節タンパク質が該感染細胞から分泌される条件下で、前記対象に投与される、態様171、175および186～188のいずれかの方法。
193. 前記疾患または病態が、炎症性または自己免疫性の疾患または病態である、態様171～175および186～192のいずれかの方法。
194. 前記疾患または病態が、抗好中球細胞質抗体（ANCA）関連血管炎、血管炎、自己免疫性皮膚疾患、移植手術（transplantation）、リウマチ性疾患、炎症性消化器疾患、炎症性眼疾患、炎症性神経系疾患、炎症性肺疾患、炎症性内分泌疾患、または自己免疫性血液学的疾患である、態様171～175および186～193のいずれかの方法。
195. 前記疾患または病態が、炎症性腸疾患、移植（transplant）、クローン病、潰瘍性大腸炎、多発性硬化症、喘息、関節リウマチ、または乾癬から選択される、態様193または態様194の方法。 VIII. Illustrative Embodiments
The embodiments provided include the following:
1. A variant CD80 polypeptide comprising an IgV domain or a specific binding fragment thereof, an IgC domain or a specific binding fragment thereof, or both, comprising SEQ ID NO: Positions based on the numbering of NO:28: 4, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 20, 21, 22, 24, 25, 27, 28, 29, 30, 31, 33, 36, 37, 38, 40, 41, 42, 43, 44, 47, 48, 50, 52, 53, 54, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 72, 74, 76, 77, 80, 81, 83, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 99, 102, 103, 104, 107, 108, 109, 110, 114, 115, 116, 117, 118, 120, 121, 122, 126, 127, 128, 129, 130, 133, 137, 140, 142, 143, 144, 148, 149, 152, 154, 160, 162, 164, 168, 169, 174, 175, 177, 178, 183, 185, 188, 190, 192, 193, or 199.
2. The variant CD80 polypeptide of embodiment 1, wherein said one or more amino acid modifications comprise one or more amino acid substitutions, insertions, or deletions.
3. The variant CD80 polypeptide of embodiment 1 or embodiment 2, wherein said unmodified CD80 is a mammalian CD80.
4. The variant CD80 polypeptide of any of embodiments 1-3, wherein said CD80 is human CD80.
5. An IgV domain or a specific binding fragment thereof; and
5. The variant CD80 polypeptide of any of embodiments 1 to 4, comprising an IgC domain or a specific binding fragment thereof.
6. The unmodified CD80
6. The variant CD80 polypeptide of any of embodiments 1-5, comprising: (i) the amino acid sequence as depicted in SEQ ID NO:28; (ii) an amino acid sequence having at least 95% sequence identity to SEQ ID NO:28; or (iii) a portion thereof, wherein the portion comprises an IgV domain or an IgC domain, or a specific binding fragment thereof.
7. The specific binding fragment of the IgV or IgC domain has a length of at least 50, 60, 70, 80, 90, 100, 110 or more amino acids;
The specific binding fragment of the IgV domain comprises at least 80% of the length of the IgV domain shown as amino acids 35 to 135 of SEQ ID NO:1; or the specific binding fragment of the IgC domain comprises at least 80% of the length of the IgC domain shown as amino acids 145 to 230 of SEQ ID NO:1.
The variant CD80 polypeptide of any of embodiments 1 to 6.
8. The variant CD80 polypeptide of any of embodiments 1-7, comprising up to 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20 amino acid modifications, optionally amino acid substitutions.
9. The variant CD80 polypeptide of any of embodiments 1-8, comprising an amino acid sequence exhibiting at least 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity to SEQ ID NO:28, or a specific binding fragment thereof.
10. A variant CD80 polypeptide of any of embodiments 1-9, which exhibits altered binding specificity for the ectodomain of CD28, PD-L1, or CTLA-4 compared to said unmodified CD80.
11. The variant CD80 polypeptide of embodiment 9, wherein said altered binding is altered binding affinity and/or altered binding selectivity.
12. The one or more amino acid substitutions

12. The variant CD80 polypeptide of any of embodiments 1 to 11, wherein said variant CD80 polypeptide is a cysteine, cysteine, arginine ...
13. The one or more amino acid substitutions are:

13. The variant CD80 polypeptide of any of embodiments 1 to 12, wherein
14. A variant CD80 polypeptide of any of embodiments 1-13, comprising an amino acid sequence as set forth in any of SEQ ID NOs:55-108, 280-346, 414-475, or a specific binding fragment thereof, or an amino acid sequence exhibiting at least 95% sequence identity to any of SEQ ID NOs:55-108, 280-346, 414-475, or a specific binding fragment thereof and containing said one or more amino acid substitutions.
15. A variant CD80 polypeptide of any of embodiments 1-14, comprising the IgV domain or a specific-binding fragment thereof.
16. The variant CD80 polypeptide of any of embodiments 1-15, wherein said IgV domain, or a specific fragment thereof, is the only CD80 portion of said variant CD80 polypeptide.
17. The variant CD80 polypeptide of any of embodiments 1-14, wherein said IgC domain, or a specific fragment thereof, is the only CD80 portion of said variant CD80 polypeptide.
18. A variant CD80 polypeptide of any of embodiments 1-17, comprising an amino acid sequence as set forth in any of SEQ ID NOs: 153-195, 347, 373-386, 476-477, or a specific binding fragment thereof, an amino acid sequence exhibiting at least 95% sequence identity to any of SEQ ID NOs: 153-195, 347, 373-386, 476-477, or a specific binding fragment thereof and containing said one or more amino acid substitutions.
19. A variant CD80 polypeptide of any of embodiments 1-17, which specifically binds to the ectodomain of CD28, PD-L1, or CTLA-4 with improved affinity compared to said unmodified CD80 polypeptide.
20. The variant CD80 polypeptide of any of embodiments 1-19, which specifically binds to the ectodomain of CD28, PD-L1, or CTLA-4 with improved selectivity compared to the binding of said unmodified CD80 to said ectodomain.
21. The variant CD80 polypeptide of embodiment 20, wherein said improved selectivity comprises a greater ratio of binding to one cognate binding partner selected from among CD28, PD-L1, and CTLA-4, to another of said cognate binding partners, as compared to the ratio of binding of the unmodified CD80 polypeptide to the one cognate binding partner, to the other of said cognate binding partners.
22. The variant CD80 polypeptide of embodiment 19 or embodiment 20, which specifically binds to the ectodomain of CD28 with improved selectivity compared to the binding of said unmodified CD80 to the ectodomain of CD28.
23. The variant CD80 polypeptide of embodiment 22, wherein the improved selectivity comprises a greater ratio of binding to CD28 to PD-L1 or CTLA-4 compared to the ratio of binding of the unmodified CD80 polypeptide to CD28 to PD-L1 or CTLA-4.
24. The variant CD80 polypeptide of embodiment 21 or embodiment 23, wherein said ratio is at least, or at least about, 1.5-fold, 2.0-fold, 3.0-fold, 4.0-fold, 5-fold, 10-fold, 15-fold, 20-fold, 30-fold, 40-fold, 50-fold, or more greater.
25. A variant CD80 polypeptide of any of embodiments 1-24, which specifically binds to the ectodomain of CD28 with improved affinity compared to said unmodified CD80 polypeptide.
26. The variant CD80 polypeptide of embodiment 19 or embodiment 25, wherein the improved affinity for the ectodomain is greater than 1.2-fold, 1.5-fold, 2-fold, 3-fold, 4-fold, 5-fold, 6-fold, 7-fold, 8-fold, 9-fold, 10-fold, 20-fold, 30-fold, 40-fold, 50-fold, or 60-fold improved compared to the binding affinity of said unmodified CD80 for said ectodomain.
27. The variant CD80 polypeptide of any of embodiments 1-26, wherein said one or more amino acid substitutions correspond to positions 12, 18, 20, 29, 31, 36, 40, 41, 43, 52, 59, 60, 63, 67, 70, 77, 81, 87, 88, 89, 90, 91, 92, 93, 107, 109, 114, 117, 118, 120, 122, 127, 130, 144, 169, 177, or 199 based on the numbering of SEQ ID NO:28.
28. The one or more amino acid substitutions are:

28. The variant CD80 polypeptide of any of embodiments 1 to 27, wherein the variant CD80 polypeptide is selected from the group consisting of: and conservative amino acid substitutions thereof.
29. The one or more amino acid substitutions are:

29. The variant CD80 polypeptide of any of embodiments 1 to 28, wherein
30. A variant CD80 polypeptide of any of embodiments 1-29, which specifically binds to the ectodomain of PD-L1 with improved affinity relative to said unmodified CD80 polypeptide.
31. The variant CD80 polypeptide of embodiment 30, wherein the improved affinity for the ectodomain is greater than 1.2-fold, 1.5-fold, 2-fold, 3-fold, 4-fold, 5-fold, 6-fold, 7-fold, 8-fold, 9-fold, 10-fold, 20-fold, 30-fold, 40-fold, 50-fold, or 60-fold improved compared to the binding affinity of the unmodified CD80 for the ectodomain.
32. A variant CD80 polypeptide of any of embodiments 1-31, which specifically binds to the ectodomain of PD-L1 with improved selectivity compared to binding of said unmodified CD80 to said ectodomain.
33. The variant CD80 polypeptide of embodiment 32, wherein the improved selectivity comprises a greater ratio of binding to PD-L1 to binding to CD28 or CTLA-4, compared to a ratio of binding of the unmodified CD80 polypeptide to PD-L1 to binding of the unmodified CD80 polypeptide to CD28 or CTLA-4.
34. The variant CD80 polypeptide of embodiment 33, wherein said ratio is at least or at least about 1.5-fold, 2.0-fold, 3.0-fold, 4.0-fold, 5-fold, 10-fold, 15-fold, 20-fold, 30-fold, 40-fold, 50-fold, or more greater.
35. The variant CD80 polypeptide of any of embodiments 1-34, wherein said one or more amino acid substitutions correspond to positions 12, 18, 29, 31, 33, 36, 38, 40, 41, 42, 43, 47, 48, 59, 64, 67, 68, 70, 77, 81, 85, 87, 88, 89, 90, 91, 92, 93, 94, 97, 104, 109, 115, 117, 118, 120, 122, 126, 130, 133, 140, 144, 148, 149, 168, 178, 183, 188, or 193 based on the numbering of SEQ ID NO:28.
36. The one or more amino acid substitutions are:

36. The variant CD80 polypeptide of any of embodiments 1 to 35, wherein the variant CD80 polypeptide is selected from the group consisting of: and conservative amino acid substitutions thereof.
37. The one or more amino acid substitutions are:

37. The variant CD80 polypeptide of any of embodiments 1 to 36, wherein
38. A variant CD80 polypeptide of any of embodiments 1-37, which specifically binds to the ectodomain of CD28 and to the ectodomain of PD-L1 with improved affinity compared to said unmodified CD80 polypeptide.
39. The variant CD80 polypeptide of any of embodiments 1 to 38, wherein said one or more amino acid substitutions correspond to positions 12, 18, 36, 40, 43, 59, 77, 88, 89, 90, 91, 92, 93, 94, 109, 118, 122, or 177 based on the numbering of SEQ ID NO:28.
40. The one or more amino acid substitutions are:

and conservative amino acid substitutions thereof.
41. The one or more amino acid substitutions are:

41. The variant CD80 polypeptide of any of embodiments 1 to 40, wherein
42. A variant CD80 polypeptide of any of embodiments 1-41, which specifically binds to the ectodomain of CTLA-4 with improved affinity compared to said unmodified CD80 polypeptide.
43. The variant CD80 polypeptide of embodiment 42, wherein the improved affinity for the ectodomain is greater than 1.2-fold, 1.5-fold, 2-fold, 3-fold, 4-fold, 5-fold, 6-fold, 7-fold, 8-fold, 9-fold, 10-fold, 20-fold, 30-fold, 40-fold, 50-fold, or 60-fold improved compared to the binding affinity of the unmodified CD80 for the ectodomain.
44. A variant CD80 polypeptide of any of embodiments 1-43, which specifically binds to the ectodomain of CTLA-4 with improved selectivity compared to the binding of said unmodified CD80 to said ectodomain.
45. The variant CD80 polypeptide of embodiment 44, wherein the improved selectivity comprises a greater ratio of binding to CTLA-4 to CD28 or PD-L1 compared to the ratio of binding of the unmodified CD80 polypeptide to CTLA-4 to CD28 or PD-L1.
46. The variant CD80 polypeptide of embodiment 45, wherein said ratio is at least or at least about 1.5-fold, 2.0-fold, 3.0-fold, 4.0-fold, 5-fold, 10-fold, 15-fold, 20-fold, 30-fold, 40-fold, 50-fold or more greater.
47. The variant CD80 polypeptide of any of embodiments 1-46, wherein said one or more amino acid substitutions correspond to positions 4, 29, 31, 36, 40, 41, 52, 67, 68, 70, 87, 88, 89, 90, 91, 92, 93, 107, 109, 110, 118, 120, 130, 144, or 169 based on the numbering of SEQ ID NO:28.
48. The one or more amino acid substitutions are:

48. The variant CD80 polypeptide of any of embodiments 1-47, wherein the variant CD80 polypeptide is selected from the group consisting of: and conservative amino acid substitutions thereof.
49. The one or more amino acid substitutions are:

49. The variant CD80 polypeptide of any of embodiments 1 to 48, wherein
50. A variant CD80 polypeptide of any of embodiments 1-49, which specifically binds to the ectodomain of CD28 and to the ectodomain of CTLA-4 with improved affinity compared to said unmodified CD80 polypeptide.
51. The variant CD80 polypeptide of any of embodiments 1-50, wherein said one or more amino acid substitutions correspond to positions 36, 40, 52, 70, 88, 89, 90, 91, 92, 93, 107, 118, 120, 130, 144, or 169 of SEQ ID NO:28.
52. The one or more amino acid substitutions are:

52. The variant CD80 polypeptide of any of embodiments 1-51, wherein the variant CD80 polypeptide is selected from the group consisting of: and conservative amino acid substitutions thereof.
53. The one or more amino acid substitutions are:

53. The variant CD80 polypeptide of any of embodiments 1 to 52, wherein
54. A variant CD80 polypeptide of any of embodiments 1-53, which specifically binds to the ectodomain of PD-L1 and to the ectodomain of CTLA-4 with improved affinity compared to said unmodified CD80 polypeptide.
55. The variant CD80 polypeptide of any of embodiments 1-54, wherein said one or more amino acid substitutions correspond to positions 29, 31, 36, 40, 41, 67, 70, 87, 88, 89, 90, 91, 92, 93, 109, 118, 120, 122, 130, or 178 of SEQ ID NO:28.
56. The one or more amino acid substitutions are:

56. The variant CD80 polypeptide of any of embodiments 1-55, wherein said variant CD80 polypeptide is selected from the group consisting of: and conservative amino acid substitutions thereof.
57. The one or more amino acid substitutions are:

57. The variant CD80 polypeptide of any of embodiments 1 to 56, wherein
58. A variant CD80 polypeptide of any of embodiments 1-57, which specifically binds to the CD28 ectodomain, the PD-L1 ectodomain, and the CTLA-4 ectodomain with improved affinity compared to said unmodified CD80 polypeptide.
59. The variant CD80 polypeptide of any of embodiments 1-58, wherein said one or more amino acid substitutions correspond to positions 36, 40, 88, 89, 90, 91, 92, or 93 of SEQ ID NO:28.
60. The variant CD80 polypeptide of any of the preceding embodiments, wherein said one or more amino acid substitutions are selected from the group consisting of K36G, L40M, E88D, K89R, D90K, A91G, F92Y, K93R, and conservative amino acid substitutions thereof.
61. The one or more amino acid substitutions are:

61. The variant CD80 polypeptide of any of embodiments 1 to 60, wherein
62. A polypeptide that specifically binds to the ectodomain of CD28 or the ectodomain of PD-L1 with improved affinity compared to the unmodified CD80 polypeptide; and
specifically binds to the ectodomain of CTLA-4 with reduced affinity compared to the unmodified CD80 polypeptide;
The variant CD80 polypeptide of any of embodiments 1 to 41.
63. The variant CD80 polypeptide of any of embodiments 1 to 41 and 62, wherein said one or more amino acid substitutions correspond to positions 29, 31, 33, 36, 38, 41, 42, 43, 47, 63, 67, 70, 81, 85, 87, 88, 89, 90, 91, 92, 93, 94, 109, 114, 118, 120, 127, 130, 144, 148, or 149 based on the numbering of SEQ ID NO:28.
64. The one or more amino acid substitutions are:

64. The variant CD80 polypeptide of any of embodiments 1 to 41, 62, and 63, selected from the group consisting of: and conservative amino acid substitutions thereof.
65. The one or more amino acid substitutions are:

65. The variant CD80 polypeptide of any of embodiments 1-41 and 62-64, wherein
66. Specifically binds to the ectodomain of CD28 with improved affinity compared to the unmodified CD80 polypeptide; and
specifically binds to the ectodomain of CTLA-4 with reduced affinity compared to the unmodified CD80 polypeptide;
The variant CD80 polypeptide of any of embodiments 1-41 and 62-65.
67. The variant CD80 polypeptide of any of embodiments 1-41 and 62-66, wherein said one or more amino acid substitutions correspond to positions 63, 70, 81, 87, 91, 114, 118, 120, 127, or 130 of SEQ ID NO:28.
68. The one or more amino acid substitutions are:

68. The variant CD80 polypeptide of any of embodiments 1-41, and 62-67, selected from the group consisting of: and conservative amino acid substitutions thereof.
69. The one or more amino acid substitutions are:

68. The variant CD80 polypeptide of any of embodiments 1-41, and 62-67, wherein
70. A CD80 polypeptide that specifically binds to the ectodomain of PD-L1 with improved affinity compared to the unmodified CD80 polypeptide; and
specifically binds to the ectodomain of CTLA-4 with reduced affinity compared to the unmodified CD80 polypeptide;
The variant CD80 polypeptide of any of embodiments 1-41 and 62-69.
71. The variant CD80 polypeptide of any of embodiments 1-41 and 62-70, wherein said one or more amino acid substitutions correspond to positions 29, 31, 33, 36, 38, 41, 42, 43, 47, 67, 70, 81, 85, 87, 88, 89, 90, 91, 92, 93, 94, 109, 118, 120, 144, 148, or 149 of SEQ ID NO:28.
72. The one or more amino acid substitutions are:

72. The variant CD80 polypeptide of any of embodiments 1-41, and 62-71, selected from the group consisting of: and conservative amino acid substitutions thereof.
73. The one or more amino acid substitutions are:

73. The variant CD80 polypeptide of any of embodiments 1-41, and 62-72, wherein
74. A CD80 polypeptide that specifically binds to the ectodomain of CD28 and the ectodomain of PD-L1 with improved affinity compared to the unmodified CD80 polypeptide; and
specifically binds to the ectodomain of CTLA-4 with reduced affinity compared to the unmodified CD80 polypeptide;
The variant CD80 polypeptide of any of embodiments 1-41 and 62-73.
75. The variant CD80 polypeptide of any of embodiments 1-41 and 62-74, wherein said one or more amino acid substitutions correspond to positions 70, 81, 87, 91, or 120 of SEQ ID NO:28.
76. The variant CD80 polypeptide of any of embodiments 1-41, and 62-75, wherein said one or more amino acid substitutions are selected from the group consisting of L70Q, Y87N, A91G, and T120S, and conservative amino acid substitutions thereof.
77. The variant CD80 polypeptide of any of embodiments 10-76, wherein the CD28 is human CD28.
78. The variant CD80 polypeptide of any of embodiments 10 to 77, wherein the PD-L1 is human PD-L1.
79. The variant CD80 polypeptide of any of embodiments 10-78, wherein the CTLA-4 is human CTLA-4.
80. The variant CD80 polypeptide of any of embodiments 1-79, wherein said binding activity is altered (increased or decreased) by more than 1.2-fold, 1.5-fold, 2-fold, 3-fold, 4-fold, 5-fold, 6-fold, 7-fold, 8-fold, 9-fold, 10-fold, 20-fold, 30-fold, 40-fold, or 50-fold relative to said unmodified CD80 polypeptide.
81. The variant CD80 polypeptide of any of embodiments 1-80, which is a soluble protein.
82. The variant CD80 polypeptide of any of embodiments 1-81, which is linked to a multimerization domain.
83. The variant CD80 polypeptide of any of embodiments 1-81, which is a multimeric, optionally a dimeric, polypeptide comprising a first variant CD80 polypeptide linked to a multimerization domain and a second variant CD80 polypeptide linked to a multimerization domain.
84. The variant CD80 polypeptide of embodiment 83, wherein the first variant CD80 polypeptide and the second variant CD80 polypeptide are the same or different.
85. The variant CD80 polypeptide of any of embodiments 82-84, wherein the multimerization domain is an Fc domain or a variant thereof having reduced effector function.
86. The variant CD80 polypeptide of any of embodiments 1 to 85, wherein the variant CD80 polypeptide is linked to a moiety that increases the biological half-life of the polypeptide.
87. The variant CD80 polypeptide of any of embodiments 1-86, wherein the variant CD80 polypeptide is linked to an Fc domain or variant thereof having reduced effector function.
88. The Fc domain is mammalian, optionally human; or the variant Fc domain comprises one or more amino acid modifications compared to an unmodified Fc domain that is mammalian, optionally human.
88. The variant CD80 polypeptide of any of embodiments 85 to 87.
89. The variant CD80 polypeptide of any of embodiments 85 to 88, wherein the Fc domain or variant thereof comprises an amino acid sequence as shown in SEQ ID NO:226 or SEQ ID NO:227, or an amino acid sequence that exhibits at least 85% sequence identity to SEQ ID NO:226 or SEQ ID NO:227.
90. The variant CD80 polypeptide of any of embodiments 82-89, which is indirectly linked via a linker.
91. The variant CD80 polypeptide of any of embodiments 1 to 90, which is a transmembrane immunomodulatory protein, further comprising a transmembrane domain linked to the extracellular domain (ECD) of said variant CD80 polypeptide, or a specific binding fragment thereof.
92. The variant CD80 polypeptide of embodiment 91, wherein said transmembrane domain comprises an amino acid sequence as set forth as residues 243-263 of SEQ ID NO:1, or a functional variant thereof exhibiting at least 85% sequence identity to residues 243-263 of SEQ ID NO:1.
93. The variant CD80 polypeptide of embodiment 91 or embodiment 92, further comprising a cytoplasmic domain linked to said transmembrane domain.
94. The variant CD80 polypeptide of embodiment 93, wherein the cytoplasmic domain comprises an amino acid sequence as set forth as residues 264-288 of SEQ ID NO:1, or a functional variant thereof exhibiting at least 85% sequence identity to residues 264-288 of SEQ ID NO:1.
95. A variant CD80 polypeptide of any of embodiments 1 to 94, which increases IFN-γ (interferon-γ) expression in an in vitro primary T cell assay, relative to said unmodified CD80.
96. A variant CD80 polypeptide of any of embodiments 1 to 95, which reduces IFN-γ (interferon-γ) expression in an in vitro primary T cell assay compared to said unmodified CD80.
97. The variant CD80 polypeptide of any of embodiments 1-96, which is deglycosylated.
98. An immune modulating polypeptide comprising the variant CD80 of any of embodiments 1 to 97 linked to a second polypeptide comprising an immunoglobulin superfamily (IgSF) domain.
99. The immune modulatory polypeptide of embodiment 98, wherein the IgSF domain has modified affinity and exhibits altered binding to one or more of its cognate binding partners compared to an unmodified or wild-type IgSF domain.
100. The immune modulatory polypeptide of embodiment 99, wherein the IgSF domain exhibits improved binding to one or more of its cognate binding partners compared to an unmodified or wild-type IgSF domain.
101. The immune modulatory polypeptide of any of embodiments 98-100, wherein said variant CD80 is a first variant CD80 and the IgSF domain of said second polypeptide is an IgSF domain from a second variant CD80 of any of embodiments 1-97, and wherein said first and said second variant CD80 are the same or different.
102. The immune modulatory polypeptide of any of embodiments 98-101, wherein the variant CD80 polypeptide is capable of specifically binding to CD28, PD-L1, or CTLA-4, and the IgSF domain of the second polypeptide is capable of binding to a cognate binding partner other than the one to which the variant CD80 polypeptide specifically binds.
103. The immune modulatory polypeptide of any of embodiments 98-102, wherein the variant CD80 polypeptide is capable of specifically binding to CD28 or PD-L1, and the IgSF domain is capable of binding to a cognate binding partner other than the one to which the variant CD80 polypeptide specifically binds.
104. The immune modulatory polypeptide of any of embodiments 98-103, wherein the variant CD80 polypeptide is capable of specifically binding to CD28 or CTLA-4, and the IgSF domain is capable of binding to a cognate binding partner other than the one to which the variant CD80 polypeptide specifically binds.
105. The immunomodulatory protein of any of embodiments 98-103, wherein the variant CD80 polypeptide is capable of specifically binding to CTLA-4 or PD-L1, and the IgSF domain is capable of binding to a cognate binding partner other than the one to which the variant CD80 polypeptide specifically binds.
106. The immunomodulatory polypeptide of any of embodiments 98-105, wherein the IgSF domain is derived from a member of the B7 family.
107. The immunomodulatory polypeptide of any of embodiments 82-90, wherein the IgSF domain is a tumor-localizing moiety that binds to a ligand expressed on a tumor.
108. The immunomodulatory polypeptide of embodiment 107, wherein the ligand is B7H6.
109. The immunomodulatory polypeptide of embodiment 107 or embodiment 108, wherein the IgSF domain is derived from NKp30.
110. The immune modulatory polypeptide of any of embodiments 98-109, wherein the IgSF domain has modified affinity and exhibits improved binding to one or more of its cognate binding partners compared to an unmodified or wild-type IgSF domain.
111. The immunomodulatory polypeptide of any of embodiments 98-110, wherein the IgSF domain is or comprises an IgV domain.
112. The immune modulating polypeptide of any of embodiments 98-111, wherein said variant CD80 polypeptide is or comprises an IgV domain.
113. The immune modulatory polypeptide of any of embodiments 98-112, comprising a multimerization domain linked to one or both of said variant CD80 polypeptide and said second polypeptide comprising an IgSF domain.
114. The immunomodulatory protein of embodiment 113, wherein the multimerization domain is an Fc domain or a variant thereof with reduced effector function.
115. The immunomodulatory polypeptide of any of embodiments 98 to 114, which is a dimer.
116. The immunomodulatory protein of embodiment 115, which is a homodimer.
117. The immunomodulatory protein of embodiment 115, which is a heterodimer.
118. A conjugate comprising a variant CD80 of any of embodiments 1-98 or an immunomodulatory polypeptide of any of embodiments 99-117, linked to a moiety.
119. The conjugate of embodiment 118, wherein the moiety is a targeting moiety that specifically binds to a molecule on the surface of a cell.
120. The conjugate of embodiment 119, wherein the targeting moiety specifically binds to a molecule on the surface of an immune cell.
121. The conjugate of embodiment 120, wherein the immune cell is an antigen-presenting cell or a lymphocyte.
122. The conjugate of embodiment 119, wherein the targeting moiety is a tumor-localizing moiety that binds to a molecule on the surface of a tumor.
123. The conjugate of any of embodiments 118 to 122, wherein the moiety is a protein, peptide, nucleic acid, small molecule, or nanoparticle.
124. The conjugate of any of embodiments 118 to 123, wherein the moiety is an antibody or an antigen-binding fragment.
125. The conjugate of any of embodiments 118-124, which is bivalent, tetravalent, hexavalent, or octavalent.
126. A nucleic acid molecule encoding a variant CD80 polypeptide of any of embodiments 1-98 or an immunomodulatory polypeptide of any of embodiments 99-117.
127. The nucleic acid molecule of embodiment 126, which is a synthetic nucleic acid.
128. The nucleic acid molecule of embodiment 126 or embodiment 127, which is a cDNA.
129. A vector comprising the nucleic acid molecule of any of embodiments 125 to 128.
130. The vector of embodiment 129, which is an expression vector.
131. The vector of embodiment 129 or embodiment 130, which is a mammalian vector or a viral vector.
132. A cell comprising the vector of any of embodiments 129 to 131.
133. The cell of embodiment 132, which is a mammalian cell.
134. The cell of embodiment 133, which is a human cell.
135. A method for producing a variant CD80 polypeptide or an immunomodulatory protein, comprising the step of introducing a nucleic acid molecule of any of embodiments 125 to 128 or a vector of any of embodiments 129 to 131 into a host cell under conditions such that the protein is expressed in the host cell.
136. The method of embodiment 135, further comprising the step of isolating or purifying the variant CD80 polypeptide or the immunomodulatory protein from the cells.
137. A method of modifying a cell to express a variant CD80 polypeptide, comprising the step of introducing into a host cell a nucleic acid molecule encoding a variant CD80 polypeptide of any of embodiments 1-98 under conditions such that the polypeptide is expressed in the host cell.
138. A modified cell expressing a variant CD80 polypeptide of any of embodiments 1 to 98, an immunomodulatory protein of any of embodiments 99 to 117, a nucleic acid molecule of any of embodiments 125 to 128, or a vector of any of embodiments 129 to 131.
139. The modified cell of embodiment 138, wherein the variant CD80 polypeptide or immunomodulatory polypeptide comprises a signal peptide.
140. The modified cell of embodiment 138 or embodiment 139, wherein the variant CD80 polypeptide or immunomodulatory polypeptide does not comprise a transmembrane domain and/or is not expressed on the surface of the cell.
141. The modified cell of any of embodiments 138-140, wherein the variant CD80 polypeptide or immunomodulatory polypeptide is secreted from the modified cell.
142. The modified cell of embodiment 138 or embodiment 139, comprising a variant CD80 polypeptide that comprises a transmembrane domain and/or is a transmembrane immunomodulatory protein of any of embodiments 91 to 97.
143. The modified cell of embodiment 138, embodiment 139, or embodiment 142, wherein the variant CD80 polypeptide is expressed on the surface of the cell.
144. The modified cell of any of embodiments 138-143, which is an immune cell.
145. The modified cell of embodiment 144, wherein the immune cell is an antigen-presenting cell (APC) or a lymphocyte.
146. The modified cell of any of embodiments 138-145, which is a primary cell.
147. The modified cell of any of embodiments 138-146, which is a mammalian cell.
148. The modified cell of any of embodiments 138 to 147, which is a human cell.
149. The modified cell of any of embodiments 145 to 148, wherein the lymphocyte is a T cell.
150. The modified cell of any of embodiments 145 to 149, wherein the APC is an artificial APC.
151. The modified cell of any of embodiments 138-150, further comprising a chimeric antigen receptor (CAR) or a modified T cell receptor.
152. An infectious agent comprising a nucleic acid molecule encoding a variant CD80 polypeptide of any of embodiments 1-97 or an immunomodulatory polypeptide of any of embodiments 98-117.
153. The infectious agent of embodiment 152, wherein the encoded variant CD80 polypeptide or immunomodulatory polypeptide does not include a transmembrane domain and/or is not expressed on the surface of a cell that expresses it.
154. The infectious agent of embodiment 152 or embodiment 153, wherein the encoded variant CD80 polypeptide or immunomodulatory polypeptide is secreted from cells expressing it.
155. The infectious agent of embodiment 152, wherein the encoded variant CD80 polypeptide comprises a transmembrane domain.
156. The infectious agent of embodiment 152 or embodiment 155, wherein the encoded variant CD80 polypeptide is expressed on the surface of a cell that expresses it.
157. The infectious agent of any of embodiments 152-156, which is a bacterium or a virus.
158. The infectious agent of embodiment 157, wherein the virus is an oncolytic virus.
159. The infectious agent of embodiment 158, wherein the oncolytic virus is an adenovirus, an adeno-associated virus, a herpes virus, a herpes simplex virus, a vesicular stomatic virus, a reovirus, a Newcastle disease virus, a parvovirus, a measles virus, a vesicular stomatitis virus (VSV), a coxsackievirus, or a vaccinia virus.
160. The infectious agent of embodiment 159, wherein the virus specifically targets and/or is dendritic cell (DC) tropic.
161. The infectious agent of embodiment 160, wherein the virus is a lentiviral vector pseudotyped with a modified Sindbis virus envelope product.
162. The infectious agent of any of embodiments 152-161, further comprising a nucleic acid molecule encoding an additional gene product capable of bringing about the death of a target cell or enhancing or potentiating an immune response.
163. The infectious agent of embodiment 162, wherein the additional gene product is selected from an anti-cancer agent, an anti-metastatic agent, an anti-angiogenic agent, an immunomodulatory molecule, an immune checkpoint inhibitor, an antibody, a cytokine, a growth factor, an antigen, a cytotoxic gene product, a pro-apoptotic gene product, an anti-apoptotic gene product, a cell matrix degrading gene, a gene for tissue regeneration, or a gene for reprogramming human somatic cells to pluripotency.
164. A pharmaceutical composition comprising a variant CD80 polypeptide of any of embodiments 1 to 97, an immunomodulatory protein of any of embodiments 98 to 117, a conjugate of any of embodiments 118 to 125, a modified cell of any of embodiments 138 to 151, or an infectious agent of any of embodiments 152 to 163.
165. The pharmaceutical composition of embodiment 164, comprising a pharma- ceutically acceptable excipient.
166. The pharmaceutical composition of embodiment 164 or 165, which is sterile.
167. An article of manufacture comprising the pharmaceutical composition of any of embodiments 164-166 in a vial.
168. The article of manufacture of embodiment 167, wherein the vial is sealed.
169. A kit comprising the pharmaceutical composition of any of embodiments 164-166 and instructions for use.
170. A kit comprising the article of manufacture of embodiments 167 and 168 and instructions for use.
171. A method of modulating an immune response in a subject, comprising administering to the subject a pharmaceutical composition of any of embodiments 164-166.
172. A method of modulating an immune response in a subject, comprising administering a modified cell of any of embodiments 138-151.
173. The method of embodiment 172, wherein the modified cells are autologous to the subject.
174. The method of embodiment 172, wherein the modified cells are allogeneic to the subject.
175. The method of any of embodiments 171-174, wherein modulating the immune response treats a disease or condition in the subject.
176. The method of any of embodiments 171-175, wherein the immune response is enhanced.
177. The method of any of embodiments 171, 175, and 176, wherein an immunomodulatory protein or conjugate comprising a variant CD80 polypeptide linked to a tumor-localizing moiety is administered to the subject.
178. The method of embodiment 177, wherein the tumor-localizing moiety is or comprises a binding molecule that recognizes a tumor antigen.
179. The method of embodiment 178, wherein the binding molecule comprises an antibody or an antigen-binding fragment thereof, or comprises a wild-type IgSF domain or a variant thereof.
180. The method of any of embodiments 171 and 175 to 179, wherein a pharmaceutical composition comprising an immunomodulatory protein of any of embodiments 107 to 117 or a conjugate of any of embodiments 118 to 125 is administered to the subject.
181. The method of any of embodiments 172-176, wherein a modified cell comprising a variant CD80 polypeptide that is a transmembrane immunomodulatory protein is administered to the subject, and/or a modified cell of any of embodiments 138, 139, and 142-151 is administered to the subject.
182. The method of any of embodiments 171, 175, and 176, wherein an infectious agent encoding a variant CD80 polypeptide that is a transmembrane immunomodulatory protein is administered to the subject, optionally under conditions where the infectious agent infects tumor cells or immune cells and the transmembrane immunomodulatory protein is expressed on the surface of the infected cells.
183. The method of any of embodiment 181 or embodiment 182, wherein the transmembrane immunomodulatory protein is a transmembrane immunomodulatory protein of any of embodiments 91 to 97.
184. The method of any of embodiments 171-183, wherein the disease or condition is a tumor or cancer.
185. The method of any of embodiments 171-184, wherein the disease or condition is selected from melanoma, lung cancer, bladder cancer, hematological malignancies, liver cancer, brain cancer, kidney cancer, breast cancer, pancreatic cancer, colon cancer, spleen cancer, prostate cancer, testicular cancer, ovarian cancer, uterine cancer, gastric cancer, musculoskeletal cancer, head and neck cancer, gastrointestinal cancer, germ cell cancer, or endocrine and neuroendocrine cancer.
186. The method of any of embodiments 171-175, wherein the immune response is reduced.
187. The method of any of embodiments 171, 175, and 186, wherein a variant CD80 polypeptide or immunomodulatory protein that is soluble is administered to the subject.
188. The method of embodiment 187, wherein the soluble polypeptide or immunomodulatory protein is an Fc fusion protein.
189. The method of any of embodiments 171, 175, and 186 to 188, wherein a pharmaceutical composition comprising a variant CD80 polypeptide of any of embodiments 1 to 90, and 95 to 97, or an immunomodulatory protein of any of embodiments 98 to 106, and 111 to 117, is administered to said subject.
190. The method of any of embodiments 172-175 and 186, wherein modified cells comprising a secretable variant CD80 polypeptide are administered to the subject.
191. The method of any of embodiments 172-175 and 186-188, wherein the modified cells of any of embodiments 138-141 and 144-151 are administered to the subject.
192. The method of any of embodiments 171, 175, and 186-188, wherein an infectious agent encoding a variant CD80 polypeptide that is a secreted immunomodulatory protein is administered to the subject, optionally under conditions where the infectious agent infects tumor cells or immune cells and the secreted immunomodulatory protein is secreted from the infected cells.
193. The method of any of embodiments 171-175 and 186-192, wherein the disease or condition is an inflammatory or autoimmune disease or condition.
194. The method of any of aspects 171-175 and 186-193, wherein the disease or condition is antineutrophil cytoplasmic antibody (ANCA) associated vasculitis, vasculitis, an autoimmune skin disease, transplantation, a rheumatic disease, an inflammatory gastrointestinal disease, an inflammatory eye disease, an inflammatory nervous system disease, an inflammatory lung disease, an inflammatory endocrine disease, or an autoimmune hematological disease.
195. The method of embodiment 193 or embodiment 194, wherein the disease or condition is selected from inflammatory bowel disease, transplant, Crohn's disease, ulcerative colitis, multiple sclerosis, asthma, rheumatoid arthritis, or psoriasis.

IX. Working Examples
The following examples are included for illustrative purposes only and are not intended to limit the scope of the invention.

Example 1
Generation of IgSF Domain Mutant DNA Constructs Example 1 describes the generation of human CD80 IgSF domain mutant DNA constructs for translation and expression on the surface of yeast as a yeast display library.

に基づいて構築物を生成した。 A. Degenerate Libraries The extracellular domain (ECD) of the wild-type human CD80 amino acid sequence as set forth in SEQ ID NO:28 (corresponding to residues 35-242 of the sequence set forth in UniProt Accession No. P33681):

The construct was generated based on

For libraries targeting specific residues for full or partial randomization with degenerate codons, DNA encoding SEQ ID NO:28 was ordered from Integrated DNA Technologies (Coralville, IA) as a series of overlapping oligonucleotides up to 80 base pairs (bp) in length. To generate a library of diverse variants of the ECD, the oligonucleotides contained the desired degenerate codons at the desired amino acid positions. Degenerate codons were generated using the algorithm at URL: rosettadesign.med.unc.edu/SwiftLib/.

In general, positions and degenerate codons to be mutated were selected from the following: Crystal structures of the target-ligand pairs of interest were used to identify ligand contact residues as well as residues at the protein interaction interface. This analysis was performed using the structure viewer available at URL: spdbv.vital-it.ch). For example, the crystal structure of CD80 bound to CTLA-4 is publicly available at URL: www.rcsb.org/pdb/explore/explore.do?structureId=1I8L, and a targeted library was designed based on the CD80::CTLA-4 interface for the selection of improved binders to CTLA-4. However, since there are no CD80 structures available for the ligands CD28 and PDL1, the same library was also used to select binders for CD28 (which binds to the same region on CD80 as CTLA-4) and PDL1 (it is not known whether PDL1 binds to the same site as CTLA-4).

The next step in library design was the alignment of human, mouse, rat and monkey CD80 sequences to identify conserved residues. Based on this analysis, conserved target residues were mutated with degenerate codons, which only identified conservative amino acid changes plus the wild-type residue. Non-conserved residues were mutated more aggressively, but also included the wild-type residue. Degenerate codons that also encoded the wild-type residue were placed to avoid over-mutagenizing the target protein. For the same reason, only a maximum of 20 positions could be targeted at one time for mutagenesis. These residues were a combination of contact and non-contact interface residues.

Oligonucleotides were dissolved in sterile water, mixed in equimolar ratios, heated to 95°C for 5 min, and slowly cooled to room temperature for annealing. PCR products were then generated using ECD-specific oligonucleotide primers that anneal to the beginning and end of the ECD, respectively. 100 ng of PCR product from the previous step was then amplified using a modified version of the pBYDS03 cloning vector (Life Technologies USA) and ECD-specific oligonucleotides that overlap by 40-50 bp outside and within the BamH1 and Kpn1 cloning sites, generating a total of 5 μg of DNA. Both polymerase chain reactions (PCRs) were performed using OneTaq 2x PCR master mix (New England Biolabs, USA). The second PCR product was purified using a PCR purification kit (Qiagen, Germany) and resuspended in sterile deionized water.

To prepare the library insert, the modified yeast display version of the vector pBYDS03 was digested with BamH1 and Kpn1 restriction enzymes (New England Biolabs, USA), and the large vector fragment was gel purified and dissolved in sterile deionized water. Electroporatable DNA for the next step was generated by mixing 12 μg of library DNA and 4 μg of linearized vector in a total volume of 50 μl of deionized and sterile water. An alternative method for generating targeted libraries was to perform site-directed mutagenesis (Multisite kit, Agilent, USA) of the target ECD using oligonucleotides containing degenerate codons. This approach was used to generate sub-libraries that targeted only specific stretches of the target protein for mutagenesis. In these cases, the sub-libraries were mixed before proceeding to the selection step. In general, library sizes ranged from 10 ⁷ to 10 ⁸ clones, except for the sub-libraries, which only ranged from 10 ⁴ to 10 ⁵ . Large libraries and sub-libraries are generated for ICOSL.

B. Random Library A random library was also constructed to identify variants of the ECD of CD80 set forth in SEQ ID NO:28 (containing the IgV domain corresponding to residues 35-135 of the sequence set forth in UniProt Accession No. P33681, with adjacent residues C-terminal to the wild-type sequence). DNA encoding the wild-type ECD was cloned between the BamH1 and Kpn1 sites of the modified yeast display vector pBYDS03 and then excised using the same restriction enzymes. The excised DNA was then mutagenized with the Genemorph II kit (Agilent, USA) to generate an average of 3-5 amino acid changes per library variant. The mutagenized DNA was then amplified by two-step PCR and further processed as described above for the targeted library.

Example 2
Introduction of DNA libraries into yeast Example 2 describes the introduction of CD80 DNA libraries into yeast. To introduce the degenerate and random library DNA into yeast, electroporation-competent cells of the yeast strain BJ5464 (ATCC.org; ATCC number 208288) were prepared and electroporated with the electroporatable DNA from the above step in a Gene Pulser II (Biorad, USA) essentially as described (Colby, DW et al. 2004 Methods Enzymology 388, 348-358). The only exception was that the transformed cells were grown in non-inducing minimal selection SCD-Leu medium to accommodate the LEU2 selection marker carried by the modified plasmid pBYDS03.

Library size was determined by plating dilutions of freshly recovered cells onto SCD-Leu agar plates and then estimating library size from the number of single colonies from platings that produced at least 50 colonies per plate. The remainder of the electroporated culture was grown to saturation, and cells from this culture were subcultured once more in the same medium to minimize the fraction of non-transformed cells. To maintain library diversity, this subculture step was performed using an inoculum containing at least 10x more cells than the calculated library size. Cells from the second saturated culture were resuspended in fresh medium containing sterile 25% (wt/vol) glycerol to a density of 10 ¹⁰ cells/ml and stored at -80°C (frozen library stock).

One liter of SCD-Leu medium consists of 14.7 grams of sodium citrate, 4.29 grams of citric acid monohydrate, 20 grams of dextrose, 6.7 grams of Difco brand nitrogen base for yeast, and 1.6 grams of synthetic dropout medium supplement for yeast without leucine. The medium was sterile filtered before use using a 0.2 μM vacuum filtration apparatus.

Library size was determined by plating dilutions of freshly harvested cells onto SCD-Leu agar plates and then estimating library size from the number of single colonies from platings that yielded at least 50 colonies per plate.

To isolate plasmids from cells containing two or more distinct library clones, a number of cells equivalent to 10 times the library size was taken from an overnight SCD-Leu culture, subcultured 1/100 into fresh SCD-Leu medium, and grown overnight. Cells from this overnight culture were resuspended in sterile 25% (wt/vol) glycerol to a density of 10 ¹⁰ cells/ml, frozen, and stored at -80°C (frozen library stocks).

Example 3
Selection of Yeast Example 3 describes the selection of yeast expressing affinity-modified variants of CD80. A number of cells equal to at least 10 times the library size was thawed from each library stock, suspended at 0.1×10 ⁶ cells/ml in non-inducing SCD-Leu medium, and grown overnight. The next day, a number of cells equal to 10 times the library size was centrifuged at 2000 RPM for 2 minutes and resuspended at 0.5×10 ⁶ cells/ml in inducing SCDG-Leu medium. One liter of SCDG-Leu induction medium consists of 5.4 grams of Na ₂ HPO ₄ , 8.56 grams of NaH ₂ PO ₄ ·H ₂ O , 20 grams of galactose, 2.0 grams of dextrose, 6.7 grams of Difco Yeast Nitrogen Base, and 1.6 grams of synthetic yeast dropout medium supplement without leucine, dissolved in water and sterilized by passing through a 0.22 μm membrane filtration device. Cultures were grown at 20° C. for 2 days to induce expression of the library proteins on the yeast cell surface.

Cells were treated with magnetic beads to reduce non-binders and enrich for all variant CD80 polypeptides capable of binding to exogenous recombinant counter-structure proteins. For example, yeast-displayed targeted or random CD80 libraries were selected separately against each of CD28, CTLA-4, and PD-L1. This was followed by 2-3 rounds of flow cytometric sorting using exogenous counter-structure protein stains to enrich for the fraction of yeast cells displaying improved binders. Magnetic bead enrichment and flow cytometric selection were essentially as described by Keith D. Miller,1 Noah B. Pefaur,2 and Cheryl L. Baird1 Current Protocols in Cytometry 4.7.1-4.7.30, July 2008.

Along with the CD80 library, the target ligand proteins were supplied by R&D Systems (USA) as follows: human rCD28.Fc (i.e., recombinant CD28-Fc fusion protein), rPDL1.Fc, and rCTLA4.Fc. Magnetic streptavidin beads were obtained from New England Biolabs, USA. For biotinylation of counter-structural proteins, a biotinylation kit (cat# 21955, Life Technologies, USA) was used. For two-color flow cytometry sorting, a Becton Dickinson FACS Aria II sorter was used. CD80 presentation levels were monitored with an anti-hemagglutinin (anti-HA) antibody labeled with Alexafluor 488 (Life Technologies, USA). Ligand-bound Fc fusion proteins rCD28.Fc, rCTLA4.Fc, and rPDL1.Fc were detected with PE-conjugated human Ig-specific goat Fab (Jackson ImmunoResearch, USA). Dual yeast were gated using forward scatter (FSC)/side scatter (SSC) parameters, and sort gates were based on higher ligand binding detected in FL4 and more limited tag expression binding in FL1.

The yeast output from the flow cytometry sort was assayed for higher specific binding affinity. The sort output yeast were grown and re-induced to express the specific IgSF affinity-modifying domain variants they encode. This population can then be compared by flow cytometry to the parent wild-type yeast lineage, or any other selected output, such as the bead output yeast population.

Importantly, the MFI of all the above F2 outputs, measured with an anti-HA tag antibody in FL1, was not increased and in some cases was decreased compared to the wild-type strain, indicating that the improved binding was not a function of increased expression on the yeast surface of the selected variants and validating the gating strategy that selected only low expressers from among those with high ligand binding.

Selected variant CD80 ECD domains were further formatted as fusion proteins and tested for binding and functional activity as follows.

Example 4
Reformatting of the selection output as Fc fusions and in various immunomodulatory protein types Example 4 describes the reformatting of the selection output as an immunomodulatory protein containing an affinity-modified (variant) extracellular domain (ECD) of CD80 fused to an Fc molecule (variant ECD-Fc fusion molecule).

Output cells from the final flow cytometry CD80 sort were grown to terminal density in SCD-Leu medium. Plasmid DNA from each output was isolated using a Yeast Plasmid DNA Isolation Kit (Zymoresearch, USA). For Fc fusions, coding DNA for the mutant target ECD was batch amplified from plasmid DNA preps using PCR primers that added suitable restriction sites for cloning into the Fc fusion vector of choice. After restriction digestion, the PCR products were ligated into the appropriate Fc fusion vector and subsequently chemically transformed into XL1 Blue strain E. coli (Agilent, USA) or NEB5α (New England Biolabs) as per the supplier's instructions. An exemplary Fc fusion vector is pFUSE-hIgG1-Fc2 (Invivogen, USA).

Dilutions of the transformation reactions were plated on LB-agar containing 100 μg/ml carbenicillin (Teknova, USA) to generate single colonies. Up to 96 colonies from each transformation were then grown to saturation overnight at 37°C in LB-broth (Teknova cat # L8112) in 96-well plates, and a small aliquot from each well was sent for DNA sequencing of the ECD insert to identify mutations in all clones. Sample preparation for DNA sequencing was performed using a protocol provided by a service provider (Genewiz; South Plainfield, NJ). After removing samples for DNA sequencing, glycerol was then added to the remaining culture to a final glycerol content of 25%, and the plates were stored at −20°C as master plates (see below) for later use. Alternatively, samples for DNA sequencing were generated by replica plating from grown liquid cultures onto solid agar plates using a disposable 96-well replicator (VWR, USA). These plates were incubated overnight to generate growth patches, and the plates were sent to Genewiz as specified by Genewiz. In some cases, resequencing was performed to verify mutations.

After identification of clones of interest from the analysis of DNA sequencing data generated in Genewiz, clones of interest were recovered from the master plate and grown individually to density in 5 ml of liquid LB-broth containing 100 μg/ml carbenicillin (Teknova, USA), and then 2 ml of each culture was used to prepare approximately 10 μg of miniprep plasmid DNA of each clone using a standard kit such as the Pureyield kit (Promega). Identification of clones of interest generally involved the following steps: First, the DNA sequence data file was downloaded from the Genewiz website. All sequences were then manually processed to begin at the start of the ECD coding region. The processed sequences were then batch translated using a suitable program available at the URL: www.ebi.ac.uk/Tools/st/emboss_transeq/. The translated sequences were then aligned using a suitable program available at the URL: multalin.toulouse.inra.fr/multalin/multalin.html.

Clones of interest were then identified using the following criteria: 1.) the same clone occurs at least twice in the alignment, and 2.) the mutation occurs at least twice in the alignment, preferably in separate clones. Clones that met at least one of these criteria were clones that were enriched by the sorting process due to improved binding.

To generate an immunomodulatory protein that is an Fc fusion protein containing the ECD of CD80 with at least one affinity-modifying domain, a coding nucleic acid molecule was generated that encodes a protein designed as follows: a variant ECD, followed by a linker of three alanines (AAA), followed by a human IgG1 Fc containing the mutations R77C, N82G, and V87C relative to the wild-type human IgG1 Fc shown in SEQ ID NO:226. Because the construct does not contain an antibody light chain capable of forming a covalent bond with cysteine, the human IgG1 Fc also contains a substitution of a cysteine residue at position 5 with a serine residue (C5S) compared to the wild-type or unmodified Fc shown in SEQ ID NO:226.

Example 5
Expression and Purification of Fc Fusions Example 5 describes high-throughput expression and purification of Fc fusion proteins containing variant ECD CD80.

Recombinant variant Fc fusion proteins were produced in the Expi293 expression system (Invitrogen, USA). 4 μg of each plasmid DNA from the previous step was added to 200 μl Opti-MEM (Invitrogen, USA) and simultaneously, 10.8 μl ExpiFectamine was added separately to another 200 μl Opti-MEM. After 5 min, 200 μl of plasmid DNA was mixed with 200 μl ExpiFectamine and the mixture was further incubated for an additional 20 min before being added to the cells. Ten million Expi293 cells were dispensed into separate wells of a 10 ml sterile conical-bottom deep 24-well growth plate (Thomson Instrument Company, USA) in a volume of 3.4 ml of Expi293 medium (Invitrogen, USA). The plates were shaken at 120 RPM for 5 days in a mammalian cell culture incubator set at 95% humidity and 8% _CO2 . After 5 days of incubation, cells were pelleted and the culture supernatant was removed.

Protein was purified from the supernatant using a high-throughput 96-well Protein A purification kit (Catalog number 45202, Life Technologies, USA) using the manufacturer's protocol. The resulting elution fraction was buffer exchanged into PBS using a Zeba 96 well spin desalting plate (Catalog number 89807, Life Technologies, USA) using the manufacturer's protocol. Purified protein was quantified using absorbance at 280 nm measured by a Nanodrop instrument (Thermo Fisher Scientific, USA), and protein purity was assessed by loading 5 μg of protein onto a NUPAGE precast polyacrylamide gel (Life Technologies, USA) under denaturing and reducing conditions, followed by gel electrophoresis. Proteins were visualized in the gel using standard Coomassie staining.

Example 6
Assessment of binding and activity of affinity matured IgSF domain-containing molecules
A. Binding to Counterstructures Expressed on Cells This example describes Fc fusion binding studies demonstrating the specificity and affinity of CD80 domain variant immunomodulatory proteins for their cognate binding partners.

To produce cells expressing the cognate binding partners, full-length mammalian surface expression constructs for each of human CD28, CTLA-4, and PD-L1 were engineered in the pcDNA3.1 expression vector (Life Technologies), which was supplied by Genscript, USA. Binding studies were performed using the Expi293F transient transfection system (Life Technologies, USA). The number of cells required for the experiment was determined, and appropriate 30 ml-scale transfections were performed using the manufacturer's recommended protocol. For each 30 ml transfection of CD28, CTLA-4, PD-L1, or mock, 75 million Expi293F cells were incubated with 30 μg of expression construct DNA and 1.5 ml of diluted ExpiFectamine 293 reagent for 48 hours, at which point cells were harvested for staining.

For staining by flow cytometry, 200,000 cells of the appropriate transient transfection or negative control were plated in 96-well round-bottom plates. The cells were spun down and resuspended in staining buffer (PBS (phosphate-buffered saline), 1% BSA (bovine serum albumin), and 0.1% sodium azide) for 20 min to block non-specific binding. Afterwards, the cells were centrifuged again and resuspended in staining buffer containing 100 nM to 1 nM variant immunomodulatory protein depending on the experiment of each candidate variant CD80 Fc protein in 50 μl. Primary staining was performed on ice for 45 min, after which the cells were washed twice in staining buffer. PE-conjugated anti-human Fc (Jackson ImmunoResearch, USA) was diluted 1:150 in 50 μl staining buffer and added to the cells and incubated for another 30 min on ice. The secondary antibody was washed out twice, cells were fixed in 4% formaldehyde/PBS, and samples were analyzed using a FACScan flow cytometer (Becton Dickinson, USA).

Mean fluorescence intensity (MFI) was calculated for each transfectant and negative parent line using Cell Quest Pro software (Becton Dickinson, USA).

B. Characterization of Biological Activity This example further describes the characterization of the biological activity of the Fc fusion variant proteins in a human primary T cell in vitro assay.

1. Mixed lymphocyte reaction (MLR)
The biological activity of soluble rCD80.Fc was tested in human mixed lymphocyte reactions (MLR). Human primary dendritic cells (DCs) were generated by culturing monocytes isolated from PBMCs (BenTech Bio, USA) with 500 U/ml rIL-4 (R&D Systems, USA) and 250 U/ml rGM-CSF (R&D Systems, USA) in Ex-Vivo 15 medium (Lonza, Switzerland) for 7 days in vitro. 10,000 mature DCs and 100,000 purified allogeneic CD4 ⁺ T cells (BenTech Bio, USA) were co-cultured with variant CD80 Fc fusion proteins and controls in a final volume of 200 μl Ex-Vivo 15 medium in 96-well round-bottom plates. On day 5, IFN-gamma secretion in culture supernatants was analyzed using the Human IFN-gamma Duoset ELISA kit (R&D Systems, USA). Optical density was measured by a VMax ELISA Microplate Reader (Molecular Devices, USA) and quantified against titrated rIFN-γ standards included in the IFN-gamma Duo-set kit (R&D Systems, USA).

2. Anti-CD3 co-fixation assay
The costimulatory bioactivity of CD80 fusion variants was determined in an anti-CD3 co-fixation assay. Mouse anti-human CD3 (OKT3, Biolegends, USA) at 1 nM or 4 nM was diluted with rCD80.Fc variant protein at 1 nM to 80 nM in PBS. This mixture was added to tissue culture-treated flat-bottom 96-well plates (Corning, USA) overnight to promote attachment of the stimulatory proteins to the wells of the plate. The next day, unbound proteins were washed off the plates and 100,000 purified human pan T cells (BenTech Bio, US) or human T cell clone BC3 (Astarte Biologics, USA) in a final volume of 200 μl of Ex-Vivo 15 media (Lonza, Switzerland) were added to each well. After culturing the cells for 3 days, culture supernatants were harvested and human IFN-γ levels were measured with the Duoset ELISA kit (R&D Systems, USA) as described above.

C. Results The results of binding and activity studies for the exemplary test variants are shown in Tables 7 and 8. Specifically, Table 7 shows exemplary IgSF domain amino acid substitutions (replacements) in the ECD of CD80 selected in affinity maturation screening against the respective cognate construct CD28. Table 8 shows exemplary IgSF domain amino acid substitutions (replacements) in the ECD of CD80 selected in affinity maturation screening against the respective cognate construct PD-L1. In the tables, the exemplary amino acid substitutions are indicated by the amino acid position number corresponding to the respective reference unmodified ECD sequence. For example, the reference unmodified ECD sequence in Tables 7 and 8 is the unmodified CD80 ECD sequence shown in SEQ ID NO:28. The amino acid position is shown in the center, the corresponding unmodified (e.g., wild-type) amino acid is shown before the number, and the identified variant amino acid substitution is shown after the number. The second column shows the SEQ ID NO of the variant ECD of each variant ECD-Fc fusion molecule.

Also shown are binding activities as measured by mean fluorescence intensity (MFI) values for binding of each variant Fc fusion molecule to cells engineered to express the cognate counter-structure ligand, and as measured by ratio compared to the MFI for binding of the corresponding unmodified ECD-Fc fusion molecule that does not contain the amino acid substitution to the counter-structure ligand expressed by the same cells. Also shown are functional activities of the variant Fc fusion molecules to modulate T cell activity, based on calculated levels (pg/ml) of IFN-γ in culture supernatants produced with either i) the indicated variant ECD-Fc fusion molecule co-immobilized with anti-CD3, or ii) the indicated variant ECD-Fc fusion molecule in an MLR assay. Tables 7 and 8 also show the ratio of IFN-γ produced by each variant ECD-Fc compared to the corresponding unmodified ECD-Fc in both functional assays.

As shown, the selection led to the identification of a number of affinity-modified CD80 IgSF domain variants that exhibited improved binding to at least one, and in some cases multiple, cognate counter-structure ligands. In addition, the results showed that the affinity modifications of the variant molecules also exerted improved activity to enhance and/or reduce immune activity depending on the format of the molecule. For example, co-immobilization of ligands likely provides multivalent interactions with cells to cluster or increases avidity favoring agonist activity to increase T cell activation compared to unmodified (e.g., wild-type) ECD-Fc molecules that do not contain amino acid exchanges. However, when the molecules were provided as bivalent Fc molecules in solution, the same IgSF domain variants exhibited antagonist activity that reduced T cell activation compared to unmodified (e.g., wild-type) ECD-Fv molecules that do not contain amino acid exchanges.

Table 7. CD80 variants selected against CD28. Molecular sequences, binding data, and costimulatory bioactivity data.

Table 8. CD80 variants selected against PD-L1. Molecular sequences, binding data, and costimulatory bioactivity data

Example 7
Ligand Binding Competition Assays As shown in Example 6, several variant CD80 molecules demonstrated improved binding to one or both of CD28 and PD-L1. To further evaluate the binding activity of CD80 to its ligands CD28 and PD-L1, this example describes a ligand competition assay to evaluate the non-competitive ability of exemplary variant CD80 polypeptides to bind to both CD28 and PD-L1.

An ELISA-based binding assay incorporating plate-bound variant CD80 A91G ECD-Fc was set up to assess the ability of CD80 to bind CD28 and PD-L1 simultaneously. Maxisorp 96-well ELISA plates (Nunc, USA) were coated overnight with 100 nM human recombinant variant CD80 A91G ECD-Fc fusion protein in PBS. The following day, unbound proteins were washed off and plates were blocked with 1% bovine serum albumin (Millipore, USA)/PBS for 1 h at room temperature. The blocking reagent was washed off three times with PBS/0.05% Tween, which included a 2 min incubation on a platform shaker between each wash.

In one set of competitive assays, CD80 was incubated with CD28, and the CD28-bound CD80 was then assayed for competitive binding in the presence of either other known CD80 ligand counter constructs PD-L1 or CTLA-4, or the negative control ligand PD-L2. Specifically, biotinylated recombinant human CD28 Fc fusion protein (rCD28.Fc; R&D Systems) was titrated in a volume of 25 μl by 8-point dilutions at 1:2 dilutions starting at 10 nM to the wells. Immediately following the addition of biotinylated rCD28.Fc, unlabeled competitive binders, namely recombinant human PD-L1 monomeric his-tagged protein, recombinant human CTLA-4 monomeric his-tagged protein, or negative control human recombinant PD-L2 Fc fusion protein (R&D Systems), were added to the wells at 2000/1000/500 nM in a volume of 25 μl, respectively, in a final volume of 50 μl. These proteins were incubated together for 1 hour, after which three washing steps were repeated as described above.

After washing, 2.5ng per well of HRP-conjugated streptavidin (Jackson Immunoresearch, USA) was diluted in 1% BSA/PBS and added to the wells to detect bound biotinylated rCD28.Fc. After 1 hour of incubation, the wells were washed again three times as above. To detect the signal, following washing, 50μl of TMB substrate (Pierce, USA) was added to the wells and incubated for 7 minutes, after which 50ul of 2M sulfuric acid stop solution was added. Optical density was determined with an Emax Plus microplate reader (Molecular Devices, USA). Optical density values were graphed with Prism (Graphpad, USA).

The results are shown in Figure 7A. The results showed reduced binding of biotinylated rCD28.Fc to variant CD80 A91G ECD-Fc fusion protein following titration of rCD28.Fc. When rCD28.Fc binding was performed in the presence of the non-competitive control protein rPDL2, there was no reduction in CD28 binding to CD80 (closed triangles). In contrast, the competitive control protein rCTLA-4, when incubated with CD28.Fc, resulted in reduced CD28 binding to CD80 as expected (x-ray). When recombinant PD-L1 was incubated with CD28.Fc, no reduction in CD28 binding to CD80 was observed, demonstrating that the epitopes of CD28 and PD-L1 for CD80 are non-competitive. Binding of the recombinant PD-L1 protein used in the CD28 competition assay to CD80 was confirmed by incubation of biotinylated PD-L1 in the presence of non-biotinylated rCD28.Fc (boxes).

A reverse competition was also set up in which CD80 was incubated with PD-L1 and then the PD-L1-bound CD80 was subsequently assayed for competitive binding in the presence of either other known CD80 ligand counter constructs CD28 or CTLA-4 or the negative control ligand PD-L2. Specifically, the assay was performed by titrating biotinylated recombinant human PD-L1-His monomeric protein into wells containing recombinant variant CD80. As binding to this ligand is weaker, titrations were performed in 25μl with eight similar 1:2 dilutions starting at 5000nM. When rPD-L1-his was used to detect binding, competing ligand human rCD28.Fc, human rCTLA-4.Fc, or human rPD-L2.Fc controls were added in 25μl at a final concentration of 2.5nM in a total volume of 50μl. Subsequent washes, detection, and OD measurements were the same as above.

The results are shown in Figure 7B. Titered PD-L1-his binding alone confirmed that PD-L1 bound to the variant CD80 A91G ECD-Fc fusion molecule immobilized on the plate (squares). When PD-L1-His binding was performed in the presence of the non-competitive control protein rPDL2, there was no reduction in PD-L1 binding to CD80 (triangles). The CD28 competitive control protein rCTLA-4 did not result in reduced PD-L1 binding to CD80 when incubated with PD-L1-his, even though CTLA-4 was competitive for CD28 (x-ray). This result further demonstrated the lack of competition between CD28 and PD-L1 for CD80 binding. Finally, when PD-L1-his was incubated with CD28.Fc, no reduction in PD-L1 binding to CD80 was observed, demonstrating that the epitopes of CD28 and PD-L1 for CD80 are non-competitive.

Thus, the results showed that CTLA-4, but not PD-L1 or the negative control PD-L2, competed for CD28 binding to CD80 (Figure 7A), and that CD28, CTLA-4, and PD-L2 did not compete for PD-L1 binding to CD80 (Figure 7B). Thus, these results demonstrated that CD28 and PD-L1 are noncompetitive binders of CD80, and that this noncompetitive binding can be demonstrated regardless of which ligand is being detected in the ELISA.

Example 8
Additional Affinity-Modified IgSF Domains This example describes the design, production, and screening of additional affinity-modified ICOSL, CD86 (B7-2), and NKp30 immunomodulatory proteins, other components of the immune synapse (IS) with demonstrated dual roles in both immune activation and inhibition. These examples demonstrate that affinity modification of IgSF domains produces proteins that can act both to enhance and reduce immune activity. This work also describes various combinations of these domains paired and fused (i.e., stacked) with variant affinity-modified CD80 to form type II immunomodulatory proteins to achieve immunomodulatory activity.

Mutant DNA constructs of human ICOSL, CD86, and IgSF domains of NKp30 for translation and expression as yeast display libraries were generated substantially as described in Example 1. For libraries targeting specific residues of the target proteins for full or partial randomization with degenerate codons, the coding DNA sequences of the extracellular domain (ECD) of human ICOSL (SEQ ID NO:32) and the extracellular domain (ECD) of NKp30 (SEQ ID NO:54) were ordered as a series of overlapping oligonucleotides of up to 80 base pairs (bp) in length from Integrated DNA Technologies (Coralville, IA). Residues were mutated by targeted mutagenesis substantially as described in Example 1. Alternatively, random libraries were constructed to identify variants of the ECD of ICOSL (SEQ ID NO:32), the ECD of CD86 (SEQ ID NO:29), and the ECD of NKp30 (SEQ ID NO:54), substantially as described in Example 1.

Degenerate and random library DNA was introduced into yeast substantially as described in Example 2 to generate a yeast library. This library was used to select for yeast expressing affinity-modified variants of ICOSL, CD86, and NKp30 substantially as described in Example 3. Cells were treated to reduce non-binders and enrich for variants of ICOSL, CD86, or NKp30 capable of binding exogenous recombinant counter-structure protein substantially as described in Example 3. This was followed by 2-3 rounds of flow cytometric sorting using exogenous counter-structure protein stains to enrich for fractions of yeast cells displaying improved binders. Magnetic bead enrichment and flow cytometric selection were essentially as described in Keith D. Miller,1 Noah B. Pefaur,2 and Cheryl L. Baird1 Current Protocols in Cytometry 4.7.1-4.7.30, July 2008.

Along with the ICOSL, CD86 and NKp30 libraries, target ligand proteins were supplied by R&D Systems (USA) as follows: human rCD28.Fc (i.e. recombinant CD28-Fc fusion protein), rCTLA4.Fc, rICOS.Fc, and rB7H6.Fc. Two-color flow cytometry was performed substantially as described in Example 3. Yeast output from flow cytometry sorting was assayed for higher specific binding affinity. Sorting output yeast were grown and re-induced to express the specific affinity-modified IgSF domain variants they encode. This population can then be compared by flow cytometry to the parent, wild-type yeast strain, or any other selected output, e.g., bead output yeast population.

For ICOSL, the second sort output (F2) was compared to the parental ICOSL yeast for binding of each rICOS.Fc, rCD28.Fc, and rCTLA4.Fc by double staining each population with anti-HA (hemagglutinin) tag expression and anti-human Fc secondary to detect ligand binding.

For the ICOSL yeast variants selected for binding to ICOS, the F2 sort output gave a mean fluorescence intensity (MFI) value of 997 when stained with 5.6 nM rICOS.Fc, whereas the MFI of the parental ICOSL lineage was measured as 397 when stained with the same concentration of rICOS.Fc. This represents a roughly 3-fold average binding improvement in this pool of F2 selected clones, and individual clones from that pool are predicted to have much better improved MFI/affinity when tested individually.

For ICOSL yeast variants selected for binding to CD28, the F2 sort output gave an MFI value of 640 when stained with 100 nM rCD28.Fc, whereas the MFI of the parental ICOSL lineage was measured as 29 when stained with the same concentration of rCD28.Fc (a 22-fold improvement). For ICOSL yeast variants selected for binding to CTLA4, the F2 sort output gave an MFI value of 949 when stained with 100 nM rCTLA4.Fc, whereas the MFI of the parental ICOSL lineage was measured as 29 when stained with the same concentration of rCTLA4.Fc (a 32-fold improvement).

For the NKp30 yeast variant selected for binding to B7-H6, the F2 sort output gave an MFI value of 533 when stained with 16.6 nM rB7H6.Fc, whereas the MFI of the parental NKp30 lineage was measured as 90 when stained with the same concentration of rB7H6.Fc (a 6-fold improvement).

Among the NKp30 variants identified were variants containing the mutations L30V/A60V/S64P/S86G relative to a position in the NKp30 extracellular domain corresponding to the position shown in SEQ ID NO:54. Among the CD86 variants identified were variants containing the mutations Q35H/H90L/Q102H relative to a position in the CD86 extracellular domain corresponding to the position shown in SEQ ID NO:29. Among the ICOSL variants identified were variants shown in Table 8 and further described below.

As with CD80, the MFI of all the above F2 outputs was not increased and in some cases was decreased compared to the wild-type strain when measured with an anti-HA antibody in FL1, suggesting that the improved binding did not correlate with increased expression on the yeast surface of the selected variants and the gating strategy we tested only selected low expressers from among those with high ligand binding.

The exemplary selection output was reformatted as an immunomodulatory protein containing an affinity-modified (variant) extracellular domain (ECD) of ICOSL fused to an Fc molecule (variant ECD-Fc fusion molecule) substantially as described in Example 4, and the Fc fusion protein was expressed and purified substantially as described in Example 5.

Binding of the exemplary ICOSL Fc fusion variants to cell-expressed counter structures was then assessed substantially as described in Example 6. To produce cells expressing their cognate binding partners, full-length mammalian surface expression constructs for each of human CD28 and ICOS were produced substantially as described in Example 6. Binding studies and flow cytometry were performed substantially as described in Example 6. In addition, the biological activity of the Fc fusion variant proteins was characterized by either a mixed lymphocyte reaction (MLR) or anti-CD3 co-fixation assay substantially as described in Example 6.

The results of binding and activity studies for exemplary test variants are shown in Table 9. Specifically, Table 9 shows exemplary IgSF domain amino acid substitutions (exchanges) in the ECDs of the respective cognate structures ICOS and ICOSL selected in affinity maturation screens against CD28. Exemplary amino acid substitutions are indicated by the amino acid position number corresponding to the respective reference unmodified ECD sequence, as follows: For example, the reference unmodified ECD sequence in Table 9 is the unmodified ICOSL ECD sequence shown in SEQ ID NO:32. The amino acid position is shown in the center, with the corresponding unmodified (e.g., wild-type) amino acid preceding the number and the identified variant amino acid substitution following the number. The second column shows the sequence number of the variant ECD of each variant ECD-Fc fusion molecule.

Also shown is the binding activity as measured by the mean fluorescence intensity (MFI) value for binding of each variant Fc fusion molecule to cells engineered to express the cognate counter-structure ligand, compared to the ratio of MFI for binding of the corresponding unmodified ECD-Fc fusion molecule that does not contain the amino acid substitution to the counter-structure ligand expressed by the same cells. Also shown is the functional activity of the variant Fc fusion molecules to modulate T cell activity, based on the calculated pg/ml of IFN-γ in culture supernatants produced with either i) the indicated variant ECD-Fc fusion molecule co-immobilized with anti-CD3 or ii) the indicated variant ECD-Fc fusion molecule in an MLR assay. The table also shows the ratio of IFN-γ produced by each variant ECD-Fc compared to the corresponding unmodified ECD-Fc in both functional assays.

As shown, the selection led to the identification of a number of affinity-modified ICOSL IgSF domain variants that exhibited improved binding to at least one, and in some cases multiple, cognate counter-structure ligands. In addition, the results showed that the affinity modifications of the variant molecules also exhibited improved activity to enhance and reduce immune activity depending on the format of the molecule. For example, co-immobilization of ligands improves T cell activation, possibly by providing multivalent interactions with cells to cluster or by improving avidity favoring agonist activity, compared to unmodified (e.g., wild-type) ECD-Fc molecules that do not contain amino acid exchanges. However, when the molecules are provided as bivalent Fc molecules in solution, the same IgSF domain variants exhibited antagonist activity to reduce T cell activation, compared to unmodified (e.g., wild-type) ECD-Fv molecules that do not contain amino acid exchanges.

^＊：親に対する比率は、WT ICOSLに対して346pg/mlのIFN-γを使用して計算された Table 9: Selected ICOSL variants against CD28 or ICOS. Molecular sequences, binding data, and costimulatory bioactivity data.

^* : Parental ratios were calculated using 346 pg/ml IFN-γ for WT ICOSL

Example 9
Generation and Evaluation of Stacked Molecules Containing Different Affinity Modified Domains This example describes additional immunomodulatory proteins that were generated as stacked constructs containing at least two different affinity modified domains derived from the identified variant CD80 polypeptides and additional variant CD80, CD86, ICOSL, and NKp30 molecules linked together and fused to Fc.

The above selected variant molecules with modified affinities for one or more counter-structure ligands were used to generate "stack" molecules (i.e., type II immunomodulatory proteins) containing two or more affinity modified IgSF domains. The stack constructs were obtained as gene blocks (Integrated DNA Technologies, Coralville, IA) encoding the stack in a format that allows its fusion to Fc by standard Gibson assembly using a Gibson assembly kit (New England Biolabs).

Coding nucleic acid molecules for all stacks were generated that encoded proteins designed as follows: a signal peptide, followed by a first variant IgV of interest, followed by a 15 amino acid linker composed of three GGGGS (G4S) motifs (SEQ ID NO: 228), followed by a second IgV of interest, followed by two GGGGS linkers (SEQ ID NO: 229), followed by three alanines (AAA), followed by human IgG1 Fc as described above. To maximize the chance of correct folding of the IgV domains in each stack, the first IgV is preceded by all residues that normally occur in the wild-type protein between this IgV and the signal peptide (leading sequence). Similarly, the first IgV is followed by all residues that normally connect the IgV to either the next Ig domain in the wild-type protein (typically the IgC domain) or, if no such second IgV domain is present, to the transmembrane domain (trailing sequence). The same design principles were applied to the second IgV domain, except that if both IgV domains were derived from the same parent protein (e.g., CD80 IgV stacked with another CD80 IgV), the linker between them was not repeated.

Table 10 shows the design of an exemplary stacked construct. The exemplary stacked molecule shown in Table 10 contains IgV domains as indicated and further leading or trailing sequences as described above. The following components are present in the table in the following order: signal peptide (SP; SEQ ID NO:225), IgV domain 1 (IgV1), trailing sequence 1 (TS1), linker 1 (LR1; SEQ ID NO:228), IgV domain 2 (IgV2), trailing sequence 2 (TS2), linker 2 (LR2; SEQ ID NO:230), and Fc domain (SEQ ID NO:226 containing C5S/R77C/N82G/V87C amino acid substitutions). In some cases, leading sequence 1 (LS1) is present between the signal peptide and IgV1, and in some cases, leading sequence 2 (LS2) is present between the linker and IgV2.

TABLE 10 Amino acid sequences (SEQ ID NOs) of components of exemplary stacked constructs

High-throughput expression and purification of variant IgV-stacked Fc fusion molecules containing various combinations of variant IgV domains from CD80, CD86, ICOSL or Nkp30 containing at least one affinity-engineered IgV domain was performed substantially as described in Example 5. The binding of variant IgV-stacked Fc fusion molecules to their respective counter-constructs and functional activity by anti-CD3 co-immobilization assay were also evaluated substantially as described in Example 6. For example, the costimulatory bioactivity of stacked IgSF Fc fusion proteins was determined in an immobilized anti-CD3 assay similar to that described above, where 4 nM anti-CD3 (OKT3, Biolegend, USA) was co-immobilized overnight with 4 nM to 120 nM human rB7-H6.Fc (R&D Systems, USA) or human rPD-L1.Fc (R&D Systems, USA) on tissue culture-treated 96-well plates (Corning, USA). The next day, unbound proteins were washed out with PBS, and 100,000 purified pan T cells were added to each well in 100 ul of Ex-Vivo 15 medium (Lonza, Switzerland). Then, stacked IgSF domains were added in a total volume of 200 ul at concentrations ranging from 8 nM to 40 nM in a volume of 100 ul. After culturing the cells for 3 days, culture supernatants were collected and human IFN-γ levels were measured with a Duoset ELISA kit (R&D Systems, USA) as described above.

The results are shown in Tables 11-14. Table 11 shows the binding and functional activity results for variant IgV-stacked Fc fusion molecules containing a NKp30 IgV domain and a CD80 or CD86 IgV domain. Table 12 shows the binding and functional activity results for variant IgV-stacked Fc fusion molecules containing a variant CD80 IgV domain and a variant CD80, CD86, or ICOSL IgV domain. Table 13 shows the binding and functional activity results for variant IgV-stacked Fc fusion molecules containing two variant CD80 IgV domains. Table 14 shows the results for variant IgV-stacked Fc fusion molecules containing a variant CD80 IgV domain and a variant ICOSL IgV domain.

For each of Tables 11-14, column 1 shows the structural organization and orientation of the stacked, affinity modified or wild type (WT) domains, starting with the amino-terminal (N-terminal) domain, followed by the central WT or affinity modified domain before the C-terminal human IgG1 Fc domain. Column 2 shows the SEQ ID NO of each IgV domain sequence contained in each "stack" molecule. Column 3 shows the binding partner for which the affinity modified stacked domains shown in column 1 were selected.

Also shown is the binding activity as measured by the mean fluorescence intensity (MFI) values for binding of each stack molecule to cells engineered to express various counter-structure ligands, and the ratio of MFI compared to binding of the corresponding stack molecule containing an unmodified IgV domain that does not contain the amino acid substitution to the counter-structure ligand expressed by the same cells. Also shown is the functional activity of the variant stack molecules to modulate T cell activity, based on the calculated levels (pg/ml) of IFN-γ in culture supernatants produced with the indicated variant stack molecule in solution and the appropriate ligand co-immobilized with anti-CD3 as described in Example 6. The table also shows the ratio of IFN-γ produced by each variant stack molecule in the co-immobilization assay compared to the corresponding unmodified stack molecule.

As shown, the results showed that it was possible to generate stacked molecules containing at least one variant IgSF domain that exhibited improved avidity with altered affinity to at least one cognate counter-structure ligand compared to the corresponding stacked molecule containing the respective unaltered (e.g., wild-type) IgV domain. In some cases, stacked molecules derived from either one or a combination of both variant IgSF domains in the molecule exhibited improved binding to multiple cognate counter-structure ligands. The results also showed that the order of the IgV domains in the stacked molecule can, in some cases, also alter the degree of improved avidity. In some cases, functional T cell activity was also altered when assessed in a targeted co-immobilization assay.

Table 11. Stacked variant IgV Fc fusion proteins containing the NKp30 IgV domain and the CD80 or CD86 IgV domain

Table 12. Stacked variant IgV Fc fusion proteins containing CD80 IgV domain and CD80 or CD86 IgV domain

Table 13. Stacked variant IgV Fc fusion proteins containing two CD80 IgV domains

Table 14. Stacked variant IgV Fc fusion proteins containing CD80 IgV domain and ICOSL IgV domain

Example 10: Generation and evaluation of engineered cells expressing transmembrane immunomodulatory proteins Engineered T cells were generated in which transmembrane immunomodulatory proteins (TIPs) containing either the variant CD80 or ICOSL affinity-modified IgSF domain-containing extracellular domains (ECDs) were co-expressed with chimeric antigen receptors (CARs). The TIPs also contained the transmembrane and cytoplasmic domains of the corresponding wild-type CD80 or ICOSL transmembrane protein sequences. The immunomodulatory activity of the engineered cells was compared to cells expressing only CAR or cells co-expressing the corresponding wild-type CD80 or ICOSL transmembrane proteins with CAR.

An exemplary CD80-TIP was a variant CD80 having an affinity-modified IgSF domain containing amino acid mutations in the IgV and IgC domains corresponding to I67T/L70Q/A91G/T120S with respect to the positions of the transmembrane and cytoplasmic domains corresponding to residues 243-288 of SEQ ID NO:1. The amino acid sequence of an exemplary CD80-TIP is shown in SEQ ID NO:241 and is encoded by the sequence of nucleotides shown in SEQ ID NO:242. The corresponding wild-type CD80 transmembrane protein has the amino acid sequence shown as amino acid residues 35-288 of SEQ ID NO:1 and is encoded by the amino acid sequence shown in SEQ ID NO:251.

An exemplary ICOSL-TIP was a variant ICOSL having an affinity-modified IgSF domain containing an ICOSL extracellular domain as set forth in SEQ ID NO:32 and an amino acid mutation in the IgV domain corresponding to N52H/I143T for positions in the transmembrane and cytoplasmic domains corresponding to residues 257-302 of SEQ ID NO:5. The amino acid sequence of an exemplary ICOSL-TIP is set forth in SEQ ID NO:243 and is encoded by the sequence of nucleotides set forth in SEQ ID NO:244. The corresponding wild-type ICOSL transmembrane protein has the amino acid sequence set forth as amino acid residues 19-302 of SEQ ID NO:5 and is encoded by the amino acid sequence set forth in SEQ ID NO:252.

TIP containing affinity-modified domains or wild-type transmembrane proteins containing the corresponding affinity-unmodified IgSF domains were co-expressed in T cells with a first-generation chimeric antigen receptor (CAR) containing a CD3ζ intracellular signaling domain. The first-generation CAR contained an scFv specific for CD19 (SEQ ID NO:245), a hinge and transmembrane domain derived from CD8 (SEQ ID NO:246) and an intracellular signaling domain derived from CD3ζ (shown in SEQ ID NO:247). The nucleotide sequence encoding the CD19 scFv-CD3ζ CAR is shown in SEQ ID NO:248, and the amino acid sequence of the CD19 scFv-CD3ζ CAR is shown in SEQ ID NO:479.

Nucleic acid molecules were generated that encoded only the CAR or one of the exemplary TIP or wild-type transmembrane proteins (SEQ ID NO:250, encoded by the sequence of nucleotides shown in SEQ ID NO:249) separated from the CAR by the self-cleaving T2A sequence. Exemplary constructs contained the nucleic acid sequences shown in Table 15. As a control, a nucleic acid construct was also generated that encoded a second generation CAR that additionally contained a CD28 costimulatory domain (CD19 scFv-CD28-CD3ζ).

(Table 15) Nucleic acid constructs

The nucleic acid molecules were individually cloned into lentiviral vectors and used to transduce T cells isolated from human PBMC samples obtained from three different healthy donors. Lentiviral particles containing the nucleic acid sequences were produced after co-transfection of HEK293 cells with the vector and lentiviral packaging constructs. Lentiviral particles were collected from the culture medium by ultracentrifugation and titrated by qRT-PCR. Human peripheral blood mononuclear cells (PBMCs) were isolated using density sedimentation from three normal blood donors. PBMCs were cultured overnight with anti-CD3 and anti-CD28 antibodies and IL-2, and then transduced with lentiviral preparations at a multiplicity of infection of 5:1. A lentiviral vector encoding a control second-generation CAR was only used to transduce cells from one donor.

After 2 weeks (14 days) of culture, cells were analyzed for cytotoxicity after co-culture with cells expressing the target antigen using an Acea Real-Time Cell Analyzer (RTCA), which measures impedance changes in culture medium in 96-well microelectronic plates (E-plates) and shows changes in cell number and morphology in a real-time plot. CD19-expressing HeLa target cells (HeLa-CD19) were seeded in 96-well E-plates and the impedance of each monolayer was monitored for 24 hours using the RTCA system. Engineered T cells were added to the wells at an effector:target ratio of 10:1 and the wells were monitored for an additional 48 hours. The results were displayed and recorded as Cell Index (CI) values resulting from the change in measured electrical impedance, and then ratio-converted by dividing the CI readings of all wells at all time points by the CI value of the individual well at the same time (baseline time) to obtain a normalized Cell Index value that represents a percentage of the value at the base line time (see Zhang et al. "Introduction to the Data Analysis of the Roche xCELLigence® System with RTCA Package." Bioconductor. May, 3, 2016, bioconductor.org/packages/devel/bioc/vignettes/RTCA/inst/doc/aboutRTCA.pdf. Accessed September 9, 2016). In this assay, a decrease in the impedance of the monolayer reflects the killing of target cells by the transduced cells.

Results showed that decreased impedance was obtained for cells expressing first-generation CARs compared to non-transduced T cells, but the magnitude of decreased impedance for cells expressing first-generation CARs was less than for cells expressing second-generation CARs. The decreased impedance in cells expressing first-generation CARs generally continued for the first 8 hours of the assay, whereas only cells expressing second-generation CARs continued to decrease impedance thereafter.

As shown in FIG. 8, in one donor, cells co-expressing TIP or the corresponding wild-type transmembrane protein with the first generation CAR each showed a greater decrease in impedance, indicating greater cytotoxic activity compared to cells expressing only the first generation CAR. Furthermore, the results showed that cytotoxic activity was greater in CAR-expressing cells co-expressing CD80-TIP or ICOSL-TIP compared to CAR-expressing cells co-expressing the corresponding wild-type CD80 or ICOSL transmembrane protein containing an affinity-unmodified IgSF domain. The observed results of these TIP-modified cells indicated that the cytotoxic activity in cells co-expressing CD80-TIP or ICOSL-TIP with the CAR exerted enhanced activity in modulating the cytotoxic immune response of antigen-specific T cells, e.g., CAR-expressing T cells.

In two other donors, cells expressing CD80-TIP did not result in a greater impedance reduction compared to cells expressing the corresponding wild-type CD80 transmembrane protein. In one donor, it was not enough to transduce the cells with the wild-type transmembrane protein construct, but in this donor, ICOS-L TIP gave the best cytotoxicity compared to the other constructs tested. In the other donor, cells expressing ICOS-L-TIP did not result in a greater impedance reduction compared to cells expressing the corresponding wild-type ICOS-L transmembrane protein. In the cells tested, all cells co-expressing either CD80-TIP, ICOSL-TIP or the corresponding wild-type transmembrane protein with CAR showed greater cytotoxic activity than cells expressing only the first generation CAR. Differences in results observed among donors may be related to differences in T cells among donors, differences in the expression levels of various modified proteins on the surface of the cells, the specific conditions used in this exemplary assay for evaluation of cell killing (e.g., evaluation of transduced cells on day 14, evaluation of single effector:target cell ratios), or other factors.

Example 11
Assessment of Binding and Activity of CD80 IgSF Domain Variants Additional ECD CD80 variants were identified and used to produce ECD-Fc fusion proteins as described in Example 5. Binding studies were performed to essentially evaluate the specificity and affinity of the CD80 domain variant immunomodulatory proteins for their cognate binding partners as described in Example 6. Exemplary results of binding of additional CD80 IgSF domain variants to cell expressed counter constructs and biological activity from anti-CD3 cofixation assays are shown in Table 16.

The exemplary amino acid substitutions shown in Table 16 are designated by the amino acid position number corresponding to the respective reference unmodified CD80 ECD sequence shown in SEQ ID NO:28. The amino acid position is shown in the center, the corresponding unmodified (e.g., wild-type) amino acid precedes the number, and the identified variant amino acid substitution follows the number. The second column shows the variant ECD sequence number of each variant ECD-Fc fusion molecule. Also shown is the binding activity as measured by the Mean Fluorescence Intensity (MFI) value for binding of each variant Fc fusion molecule to cells modified to express a cognate counter-structure ligand, and as measured by the ratio compared to the MFI for binding of the corresponding unmodified ECD-Fc fusion molecule not containing the amino acid substitution to the counter-structure ligand expressed by the same cells. Additionally, functional activity of the variant Fc fusion molecules to modulate T cell activity is demonstrated based on calculated levels of IFN-γ (pg/ml) in cell supernatants generated with the indicated variant ECD-Fc fusion molecules co-immobilized with anti-CD3 and the ratio of IFN-γ compared to the corresponding unmodified (parent) ECD-Fc. The results demonstrate altered (including improved) binding affinity and/or improved immunological activity of CD80 IgSF domain variants having altered affinity for at least one cognate counter-structure ligand.

Table 16. CD80 variants selected against CTLA-4 or PD-L1. Molecular sequences, binding data, and costimulatory bioactivity data

The present invention is not intended to be limited in scope to the specific disclosed embodiments, which are provided, for example, to illustrate various aspects of the invention. Various modifications to the described compositions and methods will become apparent from the description and teachings herein. Such variations may be practiced without departing from the true scope and spirit of the present disclosure, and are intended to be within the scope of the present disclosure.

Claims (27)

A purified homodimer of CD80-Fc fusion,
the CD80-Fc fusion comprises one polypeptide operably linked to the Fc region of an immunoglobulin,
the one polypeptide is a variant CD80 polypeptide that contains an IgV domain;
the variant CD80 polypeptide comprises the amino acid substitution V68M , based on the amino acid numbering of SEQ ID NO:28;
the variant CD80 polypeptide is a polypeptide having an amino acid sequence that exhibits at least 93% sequence identity to (i) the amino acid sequence set forth in SEQ ID NO:28, or (ii) the amino acid sequence of positions 35 to 141 of SEQ ID NO: 1 ; and
the variant CD80 polypeptide specifically binds to the ectodomain of PD-L1 with improved affinity compared to an unmodified CD80 polypeptide comprising the amino acid sequence set forth in SEQ ID NO:28.
Purified homodimer of CD80-Fc fusion. The purified homodimer of CD80-Fc fusion of claim 1, wherein the variant CD80 polypeptide binds non-competitively to the ectodomains of CD28 and PD-L1. the variant CD80 polypeptide comprising:
Amino acid substitutions
V68M/L70P/L72P/K86E,
H18L/R29D/Y31L/Q33H/K36G/M38I/T41A/M43R/M47T/V68M/E81V/L85R/K89N/A91T/F92P/K93V/R94L/T130A,
R29D/Y31L/Q33H/K36G/M38I/T41A/M43R/M47T/V68M/E81V/L85R/K89N/A91T/F92P/K93V/R94L/T130A/N149S/R190S,
V22A/R29D/Y31L/Q33H/K36G/M38I/T41A/M43R/M47T/V68M/E81V/L85R/K89N/A91T/F92P/K93V/R94L/T130A, or
V22A/R29D/Y31L/Q33H/K36G/M38I/T41A/M43R/M47T/V68M/E81V/L85R/K89N/A91T/F92P/K93V/R94L/D115G/I118T/T130A/G133D/N149S
including,
A purified homodimer of the CD80-Fc fusion of claim 1 or 2. The purified homodimer of CD80-Fc fusion according to any one of claims 1 to 3, wherein the variant CD80 polypeptide comprises the IgV domain and the IgC domain. A purified homodimer of CD80-Fc fusion according to any one of claims 1 to 3, wherein the variant CD80 polypeptide comprises only the IgV domain. the variant CD80 polypeptide comprising:
An amino acid sequence as set forth in any of SEQ ID NOs: 320, 321, 325 and 327; or
5. The purified homodimer of CD80-Fc fusion of any one of claims 1 to 4, comprising an amino acid sequence that exhibits at least 95% sequence identity to any of SEQ ID NOs: 320, 321, 325 and 327 and contains said amino acid substitutions. The purified homodimer of CD80-Fc fusion of any one of claims 1 to 6, wherein the improved affinity for the ectodomain of PD-L1 is more than 2-fold improved compared to the binding affinity of the unmodified CD80 for the ectodomain. The purified homodimer of CD80-Fc fusion according to any one of claims 1 to 7, wherein the Fc region is a variant Fc region having reduced effector function. The Fc region is selected from the group consisting of amino acid substitutions N297A, N297Q, IgG1-N297G, IgG1-L234A/L235A, IgG1-L234A/L235E/G237A, IgG1-A325A/A330S/P331S, IgG1-C226S/C229S, IgG1-C226S/C229S/E233P/L234V/L235A, IgG1-E233P/L234V/L235A/ A purified homodimer of CD80-Fc fusion according to any one of claims 1 to 8, comprising G236del/S267K, IgG1-L234F/L235E/P331S, IgG1-S267E/L328F, IgG2-V234A/G237A, IgG2-H268Q/V309L/A330S/A331S, IgG4-L235A/G237A/E318A, or IgG4-L236E. The purified homodimer of CD80-Fc fusion according to any one of claims 1 to 7, wherein the Fc region is an Fc region derived from IgG1 or a variant IgG1 Fc region having reduced effector function. The purified homodimer of CD80-Fc fusion according to claim 10, wherein the Fc region comprises an amino acid sequence set forth in SEQ ID NO:226 or an amino acid sequence exhibiting at least 85% sequence identity to SEQ ID NO:226. The purified homodimer of CD80-Fc fusion according to claim 10 or 11, wherein the variant IgG1 Fc region comprises the amino acid substitutions R292C/N297G/V302C. The purified homodimer of CD80-Fc fusion of claim 10 or 11, wherein the variant IgG1 Fc region comprises the amino acid substitutions L234A/L235E/G237A. The purified homodimer of CD80-Fc fusion according to any one of claims 1 to 7, wherein the Fc region is an IgG2 Fc region. A purified homodimer of CD80-Fc fusion according to any one of claims 1 to 7 and 14, wherein the Fc region comprises an amino acid sequence set forth in SEQ ID NO:227 or an amino acid sequence exhibiting at least 85% sequence identity to SEQ ID NO:227. The purified homodimer of CD80-Fc fusion according to any one of claims 1 to 8, wherein the Fc region is an IgG4 Fc region or a variant IgG4 Fc region containing the mutation S228P. A purified homodimer of CD80-Fc fusion according to any one of claims 1 to 16, wherein the one polypeptide is indirectly linked to the Fc region via a linker. An in vitro method for producing a purified homodimer of a CD80-Fc fusion according to any one of claims 1 to 17, comprising the step of introducing a nucleic acid or a nucleic acid molecule encoding the Fc fusion polypeptide into a host cell under conditions in which the Fc fusion polypeptide is expressed in the host cell. The method of claim 18, further comprising isolating or purifying the CD80-Fc fusion from the cells to produce a purified homodimer of the CD80-Fc fusion. A pharmaceutical composition comprising a purified homodimer of a CD80-Fc fusion according to any one of claims 1 to 17 and a pharma- ceutical acceptable excipient. An article of manufacture or kit comprising the pharmaceutical composition of claim 20. Use of the pharmaceutical composition of claim 20 in the manufacture of a medicament for modulating an immune response in a subject. The pharmaceutical composition of claim 20 for use in modulating an immune response in a subject. 23. The use of claim 22, wherein the modulation of the immune response treats a disease or condition in the subject. 24. The pharmaceutical composition of claim 23, wherein the modulation of the immune response treats a disease or condition in the subject. 25. The use of claim 24, wherein the disease or condition is cancer. 26. The pharmaceutical composition of claim 25, wherein the disease or condition is cancer.

Images